PHYSICAL PROCESSES IN NATURAL WATERS:
PPNW 2017
Hyytiälä Forestry Field Station
Finland
21-25 August 2017
List of contents
Extended abstracts (sorted according to corresponding author)
Amadori et al., “Typical seasonal transport patterns in Lake Garda”
Amani et al., “Impact of bubble plume aeration system on manganese cycling in Blagdon Reservoir,
South West, UK”
Amorin et al., “Assessments of lake indices through a comparison between tropical and subtropical
shallow lakes”
Boehrer et al., “Methane storage and ebullition from monimolimnetic waters of polluted mine pit
lake Vollert-Sued in Germany”
Bogdanov et al., “Structural and dynamical parameters of the convectively-mixed layer in a shallow
ice-covered lake”
Bouffard et al., “Boundary layer under an ice-covered lake”
Donis et al., “Decoupling the combined effects of trophic state recovery and climate change on
Lake Hallwil water column structure”
Flury et al., “Where does methane come from and where does it go? New insights in aquatic
methane research”
Folkard, “Analysis of the layer structure of thermal microstructure profiles of stratified lakes: new
insights into vertical fluxes?”
Franz et al., “Lake-atmosphere exchange of CO2 and CH4 in arctic Siberia”
Guseva et al., “Harp Lake model intercomparison experiment: focus on vertical gas transfer”
Iwata et al., “Evaluation of methane emission from a mid-latitude lake with the eddy covariance
technique”
Jäntti et al., “Methane oxidation in Lake Kuivajärvi”
Kim et al., “Chemical enhancement of CO2 at the air-water interface in eutrophic lakes of Quebec”
Kiuru et al., “Modelling the effect of changes in air temperature and carbon loading on CO 2 in a
boreal lake”
Lemckert et al., “Designing Very Shallow Water Bodies for Disinfection: Impact of Daily
Stratification/Destratification”
Lilover et al., “Comparison of ice drift characteristics derived from Eulerian and Lagrangian
measurements in the Gulf of Finland, the Baltic Sea”
López-Moreira et al., “An integrated process-based minimal model to account for the feedbacks
between ecological and physical processes in lakes”
Ma et al., “Hydrodynamic characteristics of the Xiangxi Bay in the Three Gorges Reservoir, China”
McDonald et al., “The Role of Water Level Fluctuation on GHG Dynamics in a Temperate UK
Reservoir”
McGinnis et al., “Light and hydrodynamics as key drivers behind the recent decline of Planktothrix
rubescens in a mesotrophic lake (Lake Hallwil)”
Posada-Bedoya et al., “Effect of density currents on the seasonal evolution of basin-scale internal
waves in a Tropical Andean reservoir”
Rabe, “FVCOM modelling study of physical processes in a Scottish fjordic system”
Román-Botero et al., “Vertical mixing in a tropical Andean Reservoir, Porce II”
Rostovtseva et al., “Distribution of sea water natural constituents on shelves of Black Sea and
Brazilian coast obtained remotely from board a ship”
Simoncelli et al., “On the feasibility of kinetic energy production by Daphnia diel vertical
migration”
Spank et al., “Measurement of Greenhouse Gas Emissions from Reservoirs”
Stepanenko, “Earth rotation vs. seiching in lakes: implications for one-dimensional modelling”
Vachon et al., “Coupled methane and oxygen dynamics during distinct periods of thermal stability
in a small Swiss lake (Soppensee)”
Volkov et al., “Radiatively-driven convection in a small ice-covered lake: Dynamics of velocities
and energy dissipation”
Weber et al., “Combining downstream river demands with a sustainable raw water supply from a
drinking water reservoir”
Abstracts (sorted according to corresponding author)
Cortes et al., “Flowpath and retention of snowmelt in an ice-covered Arctic lake”
Ellis, “evaluating algal-driven shifts in coastal sediment-water oxygen dynamics”
Forrest et al., “Quantifying Spatial Variability in a Deep, sub-Alpine Lake”
Hofmann et al., “Remobilization and transport of particles in the nearshore zone of Lake
Constance”
Huttula, “Physics of boreal lakes –reflections on our learning during the last decades”
Huynh, “Thermally-driven transport of dissolved methane and carbon dioxide through the water
column in a subtropical rice field”
Jansen et al., “Hydrological controls on spring carbon gas emissions from sub-arctic lakes”
Kirillin et al., “Lake classification revisited: scaling of lake seasonal stratification”
Leppäranta et al., “Melting of lake ice: measurements and modelling”
Liu, “Methane bubble growth and transport in aquatic sediments observed by micro-scale X-ray
computed tomography”
Lorke et al., “Near-surface turbulence and gas exchange velocities in shallow streams”
MacIntyre et al., “Circulation and Respiration in Ice-covered Alaskan Arctic Lakes”
Melack et al., “Carbon dioxide fluxes in tropical waters: application of a surface renewal model
based on near surface turbulence and vertical mixing”
Qu, “Riverine carbon and nitrogen & Greenhouse gases (GHGs) emissions in rivers of the Tibetan
Plateau”
Sahlée, “Lake CO2 measurements using UAV”
Verlet-Banide, “Methane outgassing observation from lake Erken”
Wain, “Contribution of high and low frequency internal waves to boundary turbulence in a lake”
Wallin et al., “High spatial variability in stream gas transfer velocity revealed by ADV derived
turbulence measurements”
Wüest et al., “Bacteria induced mixing – comparing field observations with DNS”
List of participants
EXTENDED ABSTRACTS
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Typical seasonal transport patterns in Lake Garda
M. Amadori1*, S. Piccolroaz2,1 and M. Toffolon1
Department of Civil, Environmental and Mechanical Engineering,
University of Trento, Trento, Italy
2
Institute for Marine and Atmospheric research Utrecht,
Department of Physics and Astronomy, Utrecht University, Utrecht, The Netherlands
1
*Corresponding author, e-mail marina.amadori@unitn.it
KEYWORDS
Lake Garda; atmosphere-lake interaction; numerical modelling; transport processes; lake circulations;
Introduction
Atmospheric forcing strongly affects circulation and transport processes in lakes as wind is
the major driver of water flow. It is widely recognized that disposing of a reliable
representation of the wind forcing is necessary to achieve a good simulation of lake
circulation patterns [e.g. Lemmin and D’Adamo, 1997]. However, the combination of
complex surrounding orography and lake bathymetry may hamper the simulation of both
wind field and water motion. In this respect, Lake Garda (Italy) is an extraordinary example
of both aspects and represents a relevant, yet almost entirely unexplored, case study.
In this work, a link between the typical wind-driven circulation patterns in the lake and the
complementary water mass transport is pursued through the analysis of the correlation
between wind intensity, thermal stratification, and transport, in order to provide a first attempt
towards a systematic description of the thermo-hydrodynamics of Lake Garda.
Materials and methods
Numerical experiments are performed through an off-line coupling of the hydrodynamic
model Delft3D [Lesser et al., 2004] and the atmospheric model WRF [Skamarock et al.,
2008]. The output from WRF simulations is given at 15 minutes intervals and at a spatial
resolution of 1 km over a region containing Lake Garda. Five-day hydrodynamic simulations
are run under 6 different lake and atmosphere conditions, replicating one-day wind and air
pressure fields, which are interpolated over the lake grid as space and time varying boundary
conditions. The hydrodynamic grid has a resolution varying from 130 m in the northern
narrow part to 400 m in the southern wider part. Fixed layers are built along the vertical with
increasing thickness from the top layer (1 m) to the bottom layer (50 m), and initial condition
are set using a horizontally uniform temperature profile obtained from measurements.
The analysis is carried out evaluating residual circulations defined as an average over the time
of the daily velocity field after a spin up time of 2 days, and quantifying the transport as the
volume of water passing through reference vertical cross sections. The effects of wind
forcing, stratification conditions, and Earth rotation on the development of transport processes
in the different parts of the lake and in time are investigated through a correlation analysis.
Results and discussion
Results obtained by numerical simulations show that space and time variability of the wind
field has a key role in the development of transport processes in Lake Garda. In nearly
unstratified thermal conditions and under the forcing of a synoptic uniform and constant
Fohen wind (winter), surface water is channelled between the steep shores in the northern
narrow part towards the southern wider part (Fig. 1a). Earth rotation causes a deviation of the
flow to the right relative to the wind direction and downwelling along the western shore.
Physical Processes in Natural Waters 2017
2
Under an alternating wind field (typical of the summer season) surface currents produce
residual gyre patterns (Fig. 1b and c). In the northern part of the lake the Peler morning breeze
moves water northward, while the Ora del Garda afternoon wind blows southward: such a
regular and uniformly distributed alternation of wind fields combines to the acceleration of
the right-hand side surface flows induced by Coriolis force to produce a cyclonic gyre as a
residual outcome.
Figure 1 Residual surface circulations in Lake Garda in a winter (a) and summer case (b and c); (d)
relationship between wind intensity and total lateral transport for all seasonal simulations.
Hence, both in winter and summer cases, Earth rotation strongly affects transport deviating
the surface water current to the right with respect to the along-axis wind direction.
A quadratic dependence of lateral transport as a function of wind speed holds both for winter
and summer simulations (Fig. 1d), with different proportionality coefficient, due to the effects
of thermal stratification on the thickness of the well mixed layer and on the magnitude of
eddy viscosity. Maximum cross-correlation shows that the inertia of the lake delays water
transport behind the wind forcing. Such a delay is more relevant in winter when the thickness
of the surface layer is larger due to the lower stratification, whereas hysteretic paths are drawn
in summer because of the alternating direction of wind.
REFERENCES
Lemmin,U., D'Adamo, N. (1996), Summertime winds and direct cyclonic circulation: observations from Lake
Geneva, Ann. Geophysicae, 14,1207-1220, doi:10.1007/s00585-996-1207-z.
Lesser,G., R., Roelvink, J. A., Van Kester, J. A. T. M., Stelling, G. S. (2004), Development and validation of a
three-dimensional morphological model. Coast. Eng., 51, 883-915, doi:10.1016/j.coastaleng.2004.07.014.
Skamarock, W.C., J.B. Klemp, J. Dudhia, D.O. Gill, D.M. Barker, M.G. Duda, X.-Y. Huang, W. Wang, J.G.
Powers, (2008). A description of the advanced research WRF version 3, NCAR Technical Note TN-475+STR,
125.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Impact of bubble plume aeration system on manganese cycling
in Blagdon Reservoir, South West, UK
M. Amani1*, E. Slavin1, D. Wain1, and L. Bryant1
1
Department of Architecture and Civil Engineering,
University of Bath, Bath, United Kingdom
*Corresponding author, e-mail m.amani.geshigani@bath.ac.uk
KEYWORDS
Water quality; Manganese; drinking water reservoir; bubble plumes.
EXTENDED ABSTRACT
Introduction
Within the UK, millions of £ are spent annually by drinking water utilities to address
customer complaints with taste and odor problems stemming from excess manganese, Mn,
(e.g., brownish-colored, metallic-tasting drinking water). In the South West UK, Mn is
geologic and sedimentary and its transport within drinking water supply reservoir is largely
controlled by levels of oxygen concentrations (O 2). Removal of Mn for meeting the UK
drinking water limit (50 μg/L) is often difficult and costly using conventional water treatment
processes due to complexity of Mn redox kinetics that can create challenges for water utilities.
Thus, in the UK, many water utilities are using bubble plumes to destratify the water column
and replenish O2 in the hypolimnion to improve the source water quality prior to treatment
plant intake, which ideally minimizes the level of the required treatment in the plant. O 2
concentration plays an important role to oxidize the soluble Mn in the water column to
insoluble particulate manganese dioxide (MnO 2) that precipitate to the sediment, thereby
removing the Mn from the water column and treatment plant influent.
Materials and methods
This study was performed on Blagdon Lake, which is a man-made drinking water reservoir of
1.78 km2 in surface area, created by Bristol Water utility with the maximum depth of 12 m
near the dam and average depth of 4.3 m. To tackle the high Mn and algae concentrations in
the reservoir, seven bubble plums were installed which are in operation throughout the
summer stratification period (April- October). Field work including Mn and O 2 concentration
measurements in the water column and sediment covering the spatial distribution of Mn
throughout the reservoir were performed (April- October 2017) to investigate the efficiency of
the aeration system. Sampling strategy involved YSI castaway CTD and YSI EXO1 watercolumn profiling. Water column sampling in 2 meter increments were done using Von Dorn
water sampler for (1) subsequent O2 concentration measurements via Winkler titration method
and (2) total and soluble Mn concentration measurements at the same day via inductively
coupled plasma spectroscopy; ICP-MS. Sediment samples were taken using Uwitec sediment
corer and sent to the laboratory to determine total manganese and total organic carbon TOC
(indicative of precipitated algal biomass) via inductively coupled plasma; ICP and acid
digestion.
2
Results and discussion
Despite the aeration systems, the problem with high concentrations of Mn still persist,
average of total Mn concentration near the dam was 139 μg/L and farther away at the back of
the reservoir was 36 μg/L. Results will be presented from summer 2016 and 2017.
Preliminary data suggest that while O2 remains high throughout the reservoir, increased
mixing actually results in increased levels of Mn near the aeration systems, contrary to
management goals.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Assessments of lake indices through a comparison between
tropical and subtropical shallow lakes
Lais F. Amorim1, J Rodolfo S Martins1*, Brigitte Vinçon-Leite2, Bruno Lemaire2, Phillipe Dubois2
1
2
School of Engineering, University of São Paulo, BR
LEESU Laboratoire d’Eau Environement et Systèmes Urbains École des Ponts ParisTech
*Corresponding author, e-mail scarati@usp.br
KEYWORDS
Lakes; lakes indices, high frequency measurements; lake’s hydrodynamic.
EXTENDED ABSTRACT
Introduction
Lake indices such as Lake number, Schimidt Stability, Wedderburn number and others
has been frequently used to evaluate and predict waterbody’s stratification behavior due to
meteorological and climate conditions. Small and shallow lakes are particularly affected in
terms of their water quality by onset and mixing episodes related to energy incidence and
winds. High resolution measurements of water column temperature and atmospheric variables
like air temperature, radiation, precipitation and wind components can be employed to
characterize and predict dangerous events that affect aquatic life, water supply systems
operations, recreation and others water uses. The aim of this article is to assess computed lake
indices along onset and mixing events in shallow lakes and correlate them with main
atmospheric driving forces.
Materials and methods
Two size comparable lakes in two different climate regions were considered: Lake
Créteil, situated in Paris surroundings, France (Soulignac et al., 2017), and Hedberg
Reservoir, located 10 km from the City of Sorocaba, Sao Paulo State , Brazil (Smith et al.,
2013). Both water bodies were monitored along the year 2016 using high time resolution
meteorological variables and water temperature instruments.
Lake indices (Lake number (Ln), Schimidt Stability (St), and Wedderburn number
(W) were computed for the year of 2016 using Lake Analyzer (Read et al., 2011), an
numerical application based on indices formulations as follow (Hutchinson, 1957; Thompson
e Imberger, 1980):
𝑆𝑇 =
𝑔
𝐴𝑠
𝑧
𝑆 (𝑧 +𝑧ℎ )
1/2
𝑍𝑣
ℎ ∗ 𝑠
𝐷
∫0 (𝑧 − 𝑧𝑣 )𝜌𝑧 𝐴𝑧 𝜕𝑧 ; 𝐿𝑁 = 2𝜌 𝑇𝑢2𝑒𝐴
;𝑊=
𝑔′𝑧𝑒2
𝑢∗2 𝐿𝑠
,
where g’= g . ∆𝜌/𝜌ℎ , 𝜌ℎ is the hypolimnion density and g is the gravity acceleration, 𝑧𝑒 is the
depth of the mixed layer, 𝐿𝑠 is the lake fetch length, 𝑢∗ is the water friction velocity due the
wind stress, 𝐴𝑠 is the lake area at surface, 𝐴𝑧 is the area at depth z, 𝑧𝐷 is the maximum depth
of the lake, 𝑧𝑣 is the centre volume depth, computed as the volume weighted depth, 𝑧𝑒 and 𝑧ℎ
are the depths to the top and bottom of the metalimnion.
The resulting indices were correlated with the radiation and wind speed driving forces
considering stratified and mixed conditions. A filter was applied to consider only 10% higher
values to prevent biases due to errors in the indices determinations.
Results and discussion
With the proposed correlations, wind mixing forces are crossed with the warming
energy provide by the solar radiation, both considered in the lake indices and the mixing
status. The use of those indexes is limited when a comparison between different lakes is
aimed because each one has a specific range for the indexes values during mixed and
Physical Processes in Natural Waters 2015
2
St
1500
Stratified Hedberg
Stratified Creteil
1000
Mixed Hedberg
Mixed Creteil
Radiation (W/m²)
Radiation (W/m²)
stratified situations. Thus, combining wind and radiation to specific mixing behaviour in
similar lakes makes possible to draw a forecast chart of lake type thermal behaviour.
The assessments show that for Lake Number and Wedderburn Number there’s a
significant correlation between the indices and the driving forces. Plots on Figure 1 show a
stratified band with low wind speed (0-1,7 m.s-1), a transitional band with the wind speed
between 1,8 and 3,5 m.s-1, and a mixed band with wind speed higher than 3,5 m.s-1 are
identified. For Schmidt Stability number the correlations are not so significant.
W
1500
Stratified Hedberg
Stratified Creteil
1000
Mixed Hedberg
Mixed Creteil
500
500
0
0
0
2
4
6
8
0
2
4
Radiation (W/m²)
6
8
Wind Speed (m/s)
Wind Speed (m/s)
Ln
1500
Stratified Hedberg
Transitional
Stratified
1000
Mixed
Stratified Creteil
Mixed Hedberg
Mixed Creteil
500
0
0
1
2
3
4
5
6
7
Wind Speed (m/s)
Figure 1. Correlation between the indices (Schmidt Stability – St; Wedderburn Number – W; Lake
Number - Ln), and the proposed band division for small lakes.
References
HUTCHINSON, G. E. Treatise on Limnology. 3V. V1-Geography Physics and Chemistry.
V2-Introduction to Lake Biology and Limnoplankton. V3-Limnological Botany. John Wiley
& Sons, 1957.
READ, J. S. et al. Derivation of lake mixing and stratification indices from high-resolution
lake buoy data. Environmental Modelling & Software, v. 26, n. 11, p. 1325-1336, 2011.
ISSN 1364-8152.
SMITH, W. S.; BIAGIONI, R. C.; HALCSIK, L. Fish fauna of Floresta Nacional de Ipanema,
São Paulo State, Brazil. Biota Neotropica, v. 13, p. 175-181, 2013. ISSN 1676-0603.
SOULIGNAC, F. et al. Performance Assessment of a 3D Hydrodynamic Model Using High
Temporal Resolution Measurements in a Shallow Urban Lake. Environmental Modeling &
Assessment, p. 1-14, 2017. ISSN 1420-2026.
THOMPSON, R.; IMBERGER, J. Response of a numerical model of a stratified lake to wind
stress. Proc. 2nd Int. Symp. Stratified Flow, Trondheim, Norway, 1980. p.562-570.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Methane storage and ebullition from monimolimnetic waters of
polluted mine pit lake Vollert-Sued in Germany
Bertram Boehrer1, Christin Horn1, Philipp Metzler1,2, Karen Ullrich1,3, Matthias Koschorreck1
1
Helmholtz Centre for Environmental Research−UFZ, Magdeburg, Germany
2
current address: Kirchhoff-Institute for physics, Heidelberg, Germany
3
current address: University of Amsterdam, The Netherlands
*Corresponding author, e-mail Bertram.Boehrer@ufz.de
KEYWORDS
Dissolved gas, gas pressure, meromixis, ebullition
EXTENDED ABSTRACT
Text
Fig. 1. Possible pathways for the loss of dissolved gas from the deep waters of a meromictic lake (from [2])
Reliable gas measurements from supersaturated deep waters still remain a challenge.
However, good information is mandatory to investigate the limnic carbon cycle, assess the
endangerment through limnic eruptions and evaluate a potential source of exploitable energy.
We addressed these three points in a heavily polluted mine pit lake in Germany [2]. We
quantified the ebullition of methane from deep waters and the sediment below. Exposed to
continuous percolation of gas bubbles, the deep (monimolimnetic) water had accumulated
high concentrations of gas: directly measured gas pressures indicated the proximity to
spontaneous ebullition [1]. Consequently, the possibility of a limnic eruption was assessed by
initiating a self-sustained flow through a vertical pipe. Despite the high gas pressures, the
flow was slow and the endangerment was considered low. A sampling strategy with bags was
developed to achieve a reliable measurement of gas content and gas composition in the
monimolimnion. As a result, directly measured gas pressures could be confirmed and were
nearly exclusively attributed to methane and nitrogen. Contrary to lakes that had shown
Physical Processes in Natural Waters 2017
2
limnic eruptions, carbon dioxide played a much subordinate role, and hence the driving force
for a violent outburst of gases was missing. Nevertheless the amount of dissolved methane
was remarkably high. This presentation closes with some estimates of the commercial value
of the deposit and limiting conditions for a possible exploitation.
Fig. 1. (a) Measured gas concentration in the gas bubble of samples versus depth (b) calculated gas concentration
in the lake (c) gas pressures based on gas concentration compared with direct gas pressure measurements (from
[2]).
REFERENCES
[1] Boehrer, B., von Rohden, C., Schultze, M. (2017): Physical features of meromictic lakes: stratification and
circulation. In: Gulati, R.D., Zadereev, E.S., Degermendzhi, A.G. (eds.). Ecology of meromictic lakes.
Ecological Studies 228, Springer, Berlin, Heidelberg, New York, p. 15 - 34
[2] Horn, C., Metzler, P., Ullrich, K., Koschorreck, M., Boehrer, B. (2017): Methane storage and ebullition in
monimolimnetic waters of polluted mine pit lake Vollert-Sued, Germany. Sci. Total Environ. 584-585 , 1 10
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Structural and dynamical parameters of the convectively-mixed
layer in a shallow ice-covered lake
S. Bogdanov*, S. Volkov, G. Zdorovennova, R. Zdorovennov, T. Efremova, N. Palshin and
A. Terzhevik
Northern Water Problems Institute, Karelian Research Centre, RAS, Petrozavodsk, Russia
*Corresponding author, e-mail sergey.r.bogdanov@mail.ru
KEYWORDS
Ice-covered lakes; convection; convective layer structure; convective cells dynamic.
Introduction
The onset and development of radiatively-driven convection in a shallow ice-covered
lake play an essential role in the winter genesis of boreal lake ecosystems. The theoretical
aspects of the problem remain also challenging as conjugated with some fundamental
properties of fully developed turbulence. A number of papers devoted to this phenomenon is
rather restricted (see [Mironov et al., 2002] for references), and available experimental data
are rather limited. In particular, nor the detailed description of the structure, neither the
common vision of its driving mechanisms is not yet finally established.
Materials and methods
We present the attempt to reconstruct the qualitative picture of the convective layer,
based on analysis of data on velocity fields. The experiment was carried out during 8-13 April
2016 on boreal Lake Vendyurskoe, Russia. The Aquadopp HR-profiler was used for
measuring all three velocity components in the 2-m under-ice layer.
Results and discussion
The key experimental findings are listed below.
- The progressive-vector diagram (Fig. 1) strongly indicates the presence of large scale
circulation with velocities of order 1 mm/c and rather stable direction. One cannot exclude
that this large-scale motion is the relict of marginal heating – induced circulation as was
discussed in [Huttula et al., 2010; Kirillin et al., 2015].
- The intensities of pulsations oscillate with a period close to 3-5 hours, but the shape
of oscillations are far from sinusoidal and resemble relaxation ones with piecewise-linear
fragments.
- Horizontal velocity probability density distribution on azimuthal direction clearly
reveals the distinguishable peaks with angular distance between them close to S/2.
- The radiation does not affect crucially the dynamics of horizontal velocities, while
the vertical component is suppressed by 1-2 orders in the night.
- The system trajectories on Lumley triangle (Reynolds stress invariant map) reveal
the 2D character of structures by night. During daytime, as shown in the Figure 2, the
trajectory point moves toward prolate axisymmetric limiting curve and later, in the evening –
to oblate axisymmetric one.
To explain these features we suppose that the convectively-mixed layer constitutes the
system of eddies rotating in horizontal plane with imposed large-scale circulation. The
radiation flux makes the system more complicated due to the onset of additional circulation in
a vertical plane. At night the cells survive mostly due to their 2D structure. Besides, their
turnover time is comparable with a radiation-free interval.
Physical Processes in Natural Waters 2015
2
Fig. 1. Progressive-vector diagram for different depths; 8-13 April 2016.
As for velocity jumps and piecewise-linear character of dynamical curves, they both
find the simple explanation as a result of passing cyclone-anticyclone cells sequence across
the sensor.
Taking this qualitative picture as the basis it becomes possible to estimate such
important parameters as cells size and rotation velocities. More sophisticated statistical
analysis can shed a light on the time dependence of these parameters and some details of cells
geometry.
Fig. 2. Vizualization of turbulence evolution during daytime with Reynolds stress
invariant map. April 10 2016. Depth - 1 m.
The study was supported by the Russian Foundation for Basic Research (projects 1605-00436_a).
REFERENCES
Mironov, D., et al. (2002), Radiatively driven convection in ice-covered lakes: Observations, scaling, and a
mixed layer model. J. Geophys. Res., 107(C4), doi:10.1029/2001JC000892.
Huttula, T., et al. (2010), Modelling circulation in an ice-covered lake. Estonian Journal of Earth Sciences, 59, 4,
298–309, doi: 10.3176/earth.2010.4.06.
Kirillin, G., et al. (2015), Axisymmetric circulation driven by marginal heating in ice-covered lakes, Geophys.
Res. Lett., 42, 2893–2900, doi:10.1002/2014GL062180.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Boundary layer under an ice-covered lake
D. Bouffard1*, H. N. Ulloa2, R. Zdorovennov 3, G. Zdorovennova 3, S. Volkov3, S.
Bogdanov3, A. Terzhevik3, and A. Wüest1,2
1
Eawag, Swiss Federal Institute of Aquatic Science and Technology, Surface Waters Research and Management, Kastanienbaum, Switzerland
2
Physics of Aquatic Systems Laboratory, Margaretha Kamprad Chair, Ecole Polytechnique Federale de
Lausanne, Institute of Environmental Engineering, Lausanne, Switzerland,
3
Karelian Research Center, Northern Water Problems Institute, Russian Academy of Sciences, Petrozavodsk,
Russia
*Corresponding author, e-mail damien.bouffard@eawag.ch
KEYWORDS
Ice-covered lakes; boundary layer; winter limnology.
Introduction
In late winter, the increase in daytime duration and solar radiation can set thermal
conditions for driving a convective mixed layer under ice-covered lakes (Kirillin et al. 2012).
This convective layer has been widely studied (Mironov et al. 2002; Jonas et al. 2003; Bouffard
et al. 2016), and the associated enhanced turbulence is thought to allow early life growth in this
layer by balancing the sinking rate of non-motile phytoplankton (Kelley 1997). The boundary
layer that separates the ice from the mixed layer is also the place where biological activity has
been observed (Hampton et al. 2015). The physical properties of this type of boundary layer
remain largely unexplored and strongly differ from the ocean environment case where salinity
plays a significant role. This study aims at investigating the dynamic of the under-ice boundary
layer at diurnal scale.
Materials and methods
Two field campaigns of measurements were conducted in March 2016 and 2017 in Lake
Onego (Russia) with the goal, among others, to measure the temporal evolution of the underice boundary layer. A set of 20 TR-1060 RBR temperature loggers were deployed just below
the ice with a vertical spacing of 1.5 cm (30 cm long mooring). In order to avoid contamination,
the mooring was displaced away from the drilled hole in the ice cover, and the hole was
carefully closed with ice. The loggers were recording at 1Hz during one week. A downward
looking 600 kHz RDI-Teledyne ADCP was deployed 20 m away. The ADCP recorded in pulse
coherent mode (Mode 11) with 5 cm bin size resolution from 25 cm below the ice to 8 m deep.
Additionally, the same instrument setup used by Bouffard et al. (2016) was deployed in the
observational site, including a meteorological station, long thermistor chains, hourly CTD, and
a PAR sensor chain.
Figure 1: a) Mean and standard deviation temperature profile during day (pink color) and night time (blue color).
b) 3 days time series of temperature in the under-ice boundary layer recorded at 3 (blue), 7.5 (pink), 16.5 (red)
Physical Processes in Natural Waters 2015
2
and 30 (black) cm under the ice. Note (i) daily fluctuations, (ii) high-frequency fluctuations, and (iii) day time
unstable profiles.
Results and discussion
The under-ice boundary layer (UIBL) exhibits a clear diurnal variability as shown in Figure
1. The night time-averaged UIBL profile is simply described by a diffusive thermal layer
subjects to a top ice-temperature boundary condition, 𝑇𝑖 = 0 °C, and a lower limit temperature
boundary condition, 𝑇𝑚 ≈ 0.3 °C, set in the upper convective mixed layer during the day
before. The layer shows significant variability, and we could correlate the squeezing of the layer
to an increase in the horizontal current. The day time-averaged UIBL is characterized by a
stable very thin diffusive layer (~10 mm) that separates the ice from a maximum temperature,
and is followed by an unstable ~10-20 cm thick layer. Then, the vertical temperature gradient
flattens defining the classical mixed convective layer (Figure 1). This diurnal thermal structure
can be explained as a balance between vertical diffusion and the buoyancy flux with an
exponential decay,
𝑑 2 𝑇 𝑑𝐼
𝜅
∼
,
𝑑𝑧 2 𝑑𝑧
(1)
where 𝐼(𝑧) = (𝐸0− /𝜌0 𝑐𝑝 )𝑒 𝛾𝑧 is the kinematic radiation flux, 𝐸0− is the under-ice incident solar
irradiance, 𝜌0 = 1000 kg m-3 the reference density, 𝑐𝑝 = 4.2 × 103 J kg-1 K-1 the specific heat
of water at constant temperature, 𝛾 the extinction coefficient of radiation flux through the water
column, and 𝜅, the molecular diffusivity
Assuming (1) in the energy equation, we can solve the thermal structure, 𝑇(𝑧), in the diffusive
region, by constraining the differential equation to two boundary conditions, 𝑇 = 0 °C at 𝑧 =
0, while the lower boundary condition is characterized by a fixed temperature, 𝑇 = 𝑇𝑚 > 𝑇𝑖 at
certain 𝑧 = 𝛿. By setting this second boundary condition, we allow the thermal structure to
generate changes in stability through the diffusive region.
Figure 2: theoretical thermal profile during daytime under ice-covered water and forced by a kinetic radiation
flux, 𝐼0 = (𝐸0− /𝐶𝑝 𝜌0 ) ≈ 5 × 10−6 K m s-1, and an extension coefficient, 𝛾 ≈ 1 m-1. The boundary conditions are
𝑇𝑖 (𝑧 = 0) = 0 °C and 𝑇𝑚 (𝑧𝑚 = 0.3 𝑚) = 0.32 °C.
Figure 2 shows the theoretical thermal profile under the ice cover obtained by adopting
reference values for the kinematic radiation flux, 𝐼0 = 5 × 10−6 K m s-1, the extinction
coefficient, 𝛾 ≈ 1 m-1, and temperature boundary conditions 𝑇𝑖 (𝑧 = 0) = 0 °C and 𝑇𝑚 (𝑧𝑚 =
0.3 m) = 0.32 °C. The profile is characterized by a upper stable thermal layer 𝛿𝑠 ≈ 20 cm and
a lower unstable layer, 𝛿𝑢 ≈ 10 cm. The theoretical profile predicts a maximum unstable
vertical temperature difference of 𝛥𝑇 = 𝑇𝑚𝑎𝑥 − 𝑇𝑚 ≈ 0.09 °C, over a vertical distance ℎ𝛥𝑇 ≈
0.1 m. If we define the Rayleigh number as 𝑅𝑎 = 𝛼𝑔𝛥𝑇ℎ3 /𝜅𝜈, with 𝛼 the thermal expansion
coefficient, 𝜅 the thermal diffusivity coefficient, and 𝜈 the kinematic viscosity, we obtain a
Bouffard et al.
3
value of 𝑅𝑎 ≈ 4.3 × 105, which is much higher than the canonical critical value for free
boundary conditions (stress-free rather than no-slip at the walls), 𝑅𝑎𝐶 = 675 (Rayleigh 1916).
This theoretical result shows that the unstable thermal layer leads to free convection and it
should be an active source of convective induced turbulence and mixing. Vertical mixing should
reduce 𝛥𝑇 (𝛥𝑇𝑜𝑏𝑠 ≈ 0.01 °C) but also it should reduce/squeeze the thickness of the upper stable
layer as a consequence of buoyancy flux.
REFERENCES
Bouffard, D., R. E. Zdorovennov, G. E. Zdorovennova, N. Pasche, A. Wüest, and A. Y. Terzhevik. 2016. Icecovered Lake Onega: effects of radiation on convection and internal waves. Hydrobiologia 780: 21–36.
Hampton, S. E., M. V. Moore, T. Ozersky, E. H. Stanley, C. M. Polashenski, and A. W. E. Galloway. 2015.
Heating up a cold subject: prospects for under-ice plankton research in lakes. J. Plankton Res. 37: 277–284.
doi:10.1093/plankt/fbv002
Jonas, T., A. Y. Terzhevik, D. V. Mironov, and A. Wüest. 2003. Radiatively driven convection in an ice-covered
lake investigated by using temperature microstructure technique. J. Geophys. Res. 108: 3183.
Kelley, D. E. 1997. Convection in ice-covered lakes: effects on algal suspension. J. Plankton Res. 19: 1859–
1880.
Kirillin, G., M. Leppäranta, A. Terzhevik, and others. 2012. Physics of seasonally ice-covered lakes: a review.
Aquat. Sci. 74: 659–682.
Mironov, D., A. Terzhevik, G. Kirillin, T. Jonas, J. Malm, and D. Farmer. 2002. Radiatively driven convection
in ice-covered lakes: Observations, scaling, and a mixed layer model. J. Geophys. Res. 107: 7–16.
Rayleigh, L. 1916. On the convection currents in a horizontal layer of fluid, when the higher temperature is on
the under side. Phil. Mag. 32, 389-428.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Decoupling the combined effects of trophic state recovery and
climate change on Lake Hallwil water column structure
D. Donis1*, N. Gallina1, D. Vachon1, C. Ahnlund-McElgunn1, A. Stöckli 2, B. Ibelings 1, D. F.
McGinnis1
1
Department F.-A. Forel for environmental and aquatic Sciences (DEFSE), Faculty of Science,
University of Geneva, CH 1211, Geneva, Switzerland.
2
Department of Civil Engineering, Transportation and Environment, Canton of Argovia,
Aarau, Switzerland.
*Corresponding author, e-mail: daphne.donis@unige.ch
KEYWORDS
Lake restoration; climate change; thermal stability; Secchi depth
EXTENDED ABSTRACT
Introduction
The effects and feedbacks of climate change on lake physics and ecological
functioning are still largely uncertain. Recently, Schmid et al. (2014) predicted anomalies in
lake surface equilibrium temperature in the range of 2–5°C by the end of the 21st century, as a
response to climate change. Warming of surface waters strengthens thermal stratification
(Rempfer et al., 2010), which leads to reduced mixing. Since a small light penetration has the
effect of shoaling and strengthening the thermocline, numerous authors suggests that global
warming leads to a eutrophication effect on lake ecosystems (e.g. Johnk et al., 2008).
However, interpreting these effects from long temporal meta-analysis is complicated by the
lakes changing trophic state over the evaluated timeframe (Flaim et al., 2016).
After the drastic reduction in phosphorous load in the past 30 years, many lakes
(mainly in North America and central Europe) are going through the process of reoligotrophication (Jeppensen et al., 2005). In most cases, this has led to improved water
transparency with possibly major effects on the mixing regime by 1) deepening the
thermocline depth and 2) weakening the water column stability that would both counteract the
eutrophication due to warmer temperatures.
Since becoming highly eutrophic in the 1970’s, Lake Hallwil has undergone intensive
restoration measures. Because of these measures, the light regime has dramatically improved
since around 2000. During this time, summer Secchi depths have increased from about 2 m in
1999 to 6 meters today. In this work, we will investigate the effects of changing light regime
to the lake hydrodynamics, and compare these in context of a warming climate.
Materials and methods
We analysed the response of the hydrodynamic model CE-QUAL-W2 (Cole and
Wells, 2000) to a range of Secchi depths (SD) applied to Lake Hallwil (Switzerland, 10.2
km2, maximum depth of 46.5 m) for the period April-December 2001. Climate forcing is
resolved from 1-hr resolution data from Aarau/Buch meteo station. The applied SD of 1, 2, 4,
6 m correspond to decreasing light extinctions coefficients γ of 1.7, 0.85, 0.45, 0.28 m-1
respectively estimated by γ ≈ 1.7/SD (Wetzel 1983). These Secchi depths were in the reported
ranges in Lake Hallwil during the re-oligrotrophication process from 1987 to 2016. To test the
different responses to eutrophic and oligotrophic lake scenarios to climate change, we run the
same simulations with an increase of 2 °C on the air temperature over the analysed period.
Results and discussion
Our results on the effect of light extinction on thermocline depth and surface
temperatures are consistent with earlier studies (Rinke et al., 2010, Flaim et al., 2016). We
Physical Processes in Natural Waters 2017
2
found that increasing the SD in Lake Hallwil leads to an increased mixing depth (Fig 1a).
Surprisingly, both higher and lower surface temperatures are observed at higher SD,
depending on the heating/cooling phase of the lake. During stable stratification, surface
temperatures decrease with increasing SD, while during the onset of fall turnover, surfaces
temperatures tend to be warmer with increasing SD. This is because decreased SDs are
associated with a reduced overall heat transport into the waterbody (Tanentzap et al., 2008).
We tested the stability of the thermocline at a given SD, and the effect of an air
temperature increase by 2°C. By far, the SD had the largest influence on the lake thermal
structure. Increasing the SD depth produced a significantly weaker N 2, and a deeper N2
maximum. The effect of increasing SD can actually then compensate for atmospheric
warming. Assuming the Secchi as a proxy for trophic state, we can infer that under such a
scenario (plus 2°C), a meso-oligotrophic lake (SD = 6 m) would not experience any heat
driven shoaling of thermocline nor a significant increase in thermal stability (Fig.1b).
However, even at SD of 6 m, eutrophication effects as a response to increased air
temperatures may be due to phytoplankton blooms triggered by the warmer lake surface
temperature. While mixing depth and density gradient (N2) exert important controls on
phytoplankton, feedbacks are complex to predict, as they are depending on site-specific algal
functional traits (light inhibition, buoyancy capacities). In Lake Hallwil, the thermocline
location and stability have a critical influence on the habitat of the toxic cyanobacteria P.
rubescens (See abstract Ahnlund-McElgunn et al., this issue).
Fig. 1 Lake Hallwil top 30 meters on 3 August 2001. a) Temperature profiles at increasing water
transparency. b) N2 profiles for increasing water transparencies (solid line) and for the same at an air
temperature increased by 2°C over the analysed period (April-December 2001).
REFERENCES
Cole, T. M., and S. A. Wells (2000) CE-QUAL-W2: A two-dimensional, laterally averaged, hydrodynamic and
water quality model, version 3.0, Instr. Rep. EL-00-1, U.S. Army Eng. and Res. Dev. Cent., Vicksburg.
Jeppesen, E., et al. (2005) Lake responses to reduced nutrient loading—an analysis of contemporary long-term
data from 35 case studies Freshwat. Biol. 50, 1747–1771. doi: 10.1111/j.1365-2427.2005.01415.
Johnk, K.D., Huisman, J., Sharples, J., Sommeijer, B., Visser, P.M. & Stroom, J.M., (2008), Summer heatwaves
promote blooms of harmful cyanobacteria. Global Change Biol. 14, 495–512.
Rempfer, J. et al. (2010) The effect of the exceptionally mild European winter of 2006-2007 on temperature and
oxygen profiles in lakes in Switzerland: A foretaste of the future?, Limnol. Oceanogr., 55,.
Rinke, K.,Yeates, P., Rothhaupt, K.-O., (2010), A simulation study of the feedback of phytoplankton on thermal
structure via light extinction. Fresh- water Biol. 55: 1674–1693.
Schmid, M., S. Hunziker, and A. Wüest, (2014), Lake surface temperatures in a changing climate: A global
sensitivity analysis, Clim. Change, 124(1–2), 301–315.
Tanentzap, A.J., Yan, N.D., Keller, B., Girard, R., Heneberry, J., Gunn, J.M., Hamilton, D.P. and Taylor, P.A.,
(2008), Cooling lakes while the world warms: effects of forest regrowth and increased dissolved organic
matter on the thermal regime of a temperate, urban lake. Limnol. Oceanogr. 53, 404–410.
Wetzel, R. G. Limnology. Saunders College Publishing, Philadelphia, London, Toronto: 767 pp (1983).
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Where does methane come from and where does it go? New
insights in aquatic methane research
S. Flury1* and T. I. Battin1
1
Stream Biofilm and Ecosystem Research Laboratory, École Polytechnique Fédéral de
Lausanne, Lausanne, Switzerland,
*Corresponding author, e-mail sabine.flurymcginnis@epfl.ch
KEYWORDS
Methane; lakes; headwater streams; terrestrial-aquatic coupling; DOC quality.
EXTENDED ABSTRACT
Overview
Realizing the effect of methane (CH4) on climate change pushed scientific studies about
CH4 in aquatic water bodies forward and publications tripled from about 100 to 300 per year
during the past 10 years. CH4 is a potent greenhouse gas and is about 34x stronger than CO 2
on a 100 year basis [IPCC, 2013]. Methane production is mostly dependent on the amount
and the freshness of organic matter available for degradation [Flury et al., 2016] and it is
produced at the end of the redox chain during organic matter degradation. In aquatic systems
it is known to be produced mainly in sediments under anoxic conditions from either acetate or
carbon dioxide (CO2) as the final electron acceptors. Once, enough CH4 has accumulated and
the total gas pressure exceeds the hydrostatic pressure it may accumulate as free gas in the
sediment. Besides diffusing out of the sediment (Fig. 1), pressure drops and shear stress can
result in massive bubble release from the sediment [Joyce and Jewell, 2003; Maeck et al.,
2014] [up to 240 mmol m-2 d-1 in an impounded river - Maeck et al., 2014] and transform
aquatic waterbodies into strong greenhouse gas sources. Another effective CH4 release
pathway from sediments in shallow areas is through advective transport of CH4 gas in aquatic
vascular plants (Fig. 1). This pathway can account for a substantial fraction of CH4 emissions
from aquatic systems [Flury et al., 2010; Sorrell and Boon, 1994]. However the relative
importance of the emission pathways, strongly depends on the biomass and density of the
macrophytes [e.g. Cheng et al., 2007; Grünfeld and Brix, 1999] and on the physical properties
(e.g. lake size, depth, pressure fluctuations) of the system [Bastviken et al., 2004; Maeck et
al., 2014; McGinnis et al., 2006]. Diffusive emissions from the water column into the
atmosphere have been suggested to be exacerbated by microbubbles [McGinnis et al., 2015;
Prairie and del Giorgio, 2013] (Fig. 1). Furthermore, very recent studies suggest a
considerable contribution to diffusive CH4 emissions into the atmosphere from CH4 that is
apparently produced in lake surface layers under oxic conditions [Bogard et al., 2014;
Grossart et al., 2011; Tang et al., 2016] (Fig. 1). There are a few theories how the presence of
rather high CH4 concentrations under oxic conditions could be explained: a) CH4 is produced
in anoxic micro-niches [Grossart et al., 2011; Oremland, 1979], from algal metabolites
[Lenhart et al., 2016] or as a by-product of methylphosphonate decomposition by P-starved
bacteria [Karl et al., 2008; Yao et al., 2016] or its oxidation is inhibited by light [Murase and
Sugimoto, 2005]. Although no clear answer could yet be provided to that paradox, the oxic
methane occurrence was recognized as an additional potential CH4 source from lakes placed
close to the water-atmosphere interface [Tang et al., 2016].
Physical Processes in Natural Waters 2015
2
While knowledge about the methane cycling is rather advanced for lentic systems it is
still in its infancy for rivers and especially in headwater streams. In larger lakes organic
carbon originates mainly from internal aquatic primary production, while the terrestrialaquatic coupling is much more pronounced in streams [Brett et al., 2017]. Thus stream
metabolism is heavily subsidized by the organic carbon from the adjacent terrestrial areas.
The quantity and quality of organic matter transported to the streams largely depends on the
land-use [Brett et al., 2017]. For example it is well known that soil compaction by animal
trampling reduces soil percolation, but increases surface run-off [Alderfer and Robinson,
1947; Pietola et al., 2005]. The latter is additionally affected by the vegetation [Russell and
Bisinger, 2015].
In a recent study we investigated the controlling factors for CH 4 cycling in headwater
streams under the influence of different cattle grazing intensities in the Swiss Alps. In
contrary to our expectations we observed reduced yields of dissolved organic carbon and
dissolved inorganic nitrogen with increased grazing intensities. Furthermore, grazing
intensities also affected some of the DOC quality indices towards more fresh/autochthonous
produced organic matter (BIX and Coble peak ratio C:T). PCA analysis suggests that CH4
loads are partly controlled by delivery of organic carbon and nutrients to the streams and by
DOC qualities; however the lower the quality the higher the CH 4 load seemed to be (Fig. 2).
CH4 concentrations in these alpine streams were high and in the range of surface water
concentrations in lakes (0.1-11 umol L-1) despite oxygen saturation. A back on the envelope
estimate suggests that CH4 outgassing from headwater streams can be in the order of
hydropower dams from lowland rivers and lakes, thus revealing their potential as strong
greenhouse gas emitters. Furthermore, the occurrence of high CH4 concentrations in
headwater streams under oxic conditions reveals a similar methane paradox as is currently
under heavy debate for lakes.
Fig. 1. Schematics of emission pathways including the oxic methane production paradox (figure
taken from Tang et al. 2016).
3
Arthur et al.
-0.5
5
DOC quality (PC2 - 27%)
4
3
0.0
0.5
1.0
1.5
2.0
2.5
SUVA254
hix
CP C:T
0.5
DOC load
CH4 load
(mmol s-1)
0
2
2
NO3 load
NH4 load
1
0
4
6
0.0
8
10
-1
12
-2
14
16
-3
18
-4
-0.5
bix
-5
-6
0
5
10
delivery of ressources (PC1 - 42%)
Fig. 2. PCA analysis of driving factors for methane occurrence in 16 headwater streams in
the Swiss Alps (own data).
REFERENCES
Alderfer, R. B., and R. R. Robinson (1947), Runoff from pastures in relation to grazing intensity and soil
compaction, J. .Am. Soc. Agron., 39(11), 948-958.
Bastviken, D., J. Cole, M. Pace, and L. Tranvik (2004), Methane emissions from lakes: Dependence of lake
characteristics, two regional assessments, and a global estimate, Glob. Biogeochem. Cycle, 18(4).
Bogard, M. J., P. A. del Giorgio, L. Boutet, M. C. G. Chaves, Y. T. Prairie, A. Merante, and A. M. Derry (2014),
Oxic water column methanogenesis as a major component of aquatic CH4 fluxes, Nat. Commun., 5, 9.
Brett, M. T., et al. (2017), How important are terrestrial organic carbon inputs for secondary production in
freshwater ecosystems?, Freshwat. Biol., 62(5), 833-853.
Cheng, X., R. Peng, J. Chen, Y. Luo, Q. Zhang, S. An, J. Chen, and B. Li (2007), CH4 and N2O emissions from
Spartina alterniflora and Phragmites australis in experimental mesocosms, Chemosphere, 68, 420-427.
Flury, S., D. F. McGinnis, and M. O. Gessner (2010), Methane emissions from a freshwater marsh in response to
experimentally simulated global warming and nitrogen enrichment, J. Geophys. Res.-Biogeosciences, 115.
Flury, S., H. Roy, A. W. Dale, H. Fossing, Z. Toth, V. Spiess, J. B. Jensen, and B. B. Jorgensen (2016), Controls
on subsurface methane fluxes and shallow gas formation in Baltic Sea sediment (Aarhus Bay, Denmark),
Geochimica Et Cosmochimica Acta, 188, 297-309.
Grossart, H.-P., K. Frindte, C. Dziallas, W. Eckert, and K. W. Tang (2011), Microbial methane production in
oxygenated water column of an oligotrophic lake, P. . Natl Acad. Sci. USA, 108(49), 19657-19661.
Grünfeld, S., and H. Brix (1999), Methanogenesis and methane emissions: effects of water table, substrate type
and presence of Phragmites australis, Aquat. Bot., 64(1), 63-75.
IPCC (2013), The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of
the Intergovernmental Panel on Climate Change, in Climate Change 2013 edited by T. F. Stocker, D. Qin,
G.-K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P. M. Midgley,
Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.
Joyce, J., and P. W. Jewell (2003), Physical controls on methane ebullition from reservoirs and lakes, Environ.
Eng. Geosci., 9(2), 167-178.
Karl, D. M., L. Beversdorf, K. M. Bjorkman, M. J. Church, A. Martinez, and E. F. DeLong (2008), Aerobic
production of methane in the sea, Nature Geoscience, 1(7), 473-478.
Physical Processes in Natural Waters 2015
4
Lenhart, K., T. Klintzsch, G. Langer, G. Nehrke, M. Bunge, S. Schnell, and F. Keppler (2016), Evidence for
methane production by the marine algae Emiliania huxleyi, Biogeosciences, 13(10), 3163-3174.
Maeck, A., H. Hofmann, and A. Lorke (2014), Pumping methane out of aquatic sediments - ebullition forcing
mechanisms in an impounded river, Biogeosciences, 11(11), 2925-2938.
McGinnis, D. F., J. Greinert, Y. Artemov, S. E. Beaubien, and A. Wüest (2006), Fate of rising methane bubbles
in stratified waters: How much methane reaches the atmosphere?, J. Geophys. Res.-Oceans, 111(C9).
McGinnis, D. F., G. Kirillin, K. W. Tang, S. Flury, P. Bodmer, C. Engelhardt, P. Casper, and H. P. Grossart
(2015), Enhancing Surface Methane Fluxes from an Oligotrophic Lake: Exploring the Microbubble
Hypothesis, Environ. Sci. Technol., 49(2), 873-880.
Murase, J., and A. Sugimoto (2005), Inhibitory effect of light on methane oxidation in the pelagic water column
of a mesotrophic lake (Lake Biwa, Japan), Limnol. Oceanogr., 50(4), 1339-1343.
Oremland, R. S. (1979), Methanogenic activity in plankton samples and fish intestines - mechanism for insitu
methanogenesis in oceanic surface waters, Limnol. Oceanogr., 24(6), 1136-1141.
Pietola, L., R. Horn, and M. Yli-Halla (2005), Effects of trampling by cattle on the hydraulic and mechanical
properties of soil, Soil Tillage Res., 82(1), 99-108.
Prairie, Y. T., and P. A. del Giorgio (2013), A new pathway of freshwater methane emissions and the putative
importance of microbubbles, Inland Waters, 3(3), 311-320.
Russell, J. R., and J. J. Bisinger (2015), FORAGES AND PASTURES SYMPOSIUM: Improving soil health and
productivity on grasslands using managed grazing of livestock, J. Anim. Sci., 93(6), 2626-2640.
Sorrell, B. K., and P. I. Boon (1994), Convective gas-flow in Eleocharis sphacelata R. Br.: Methane transport
and release from wetlands, Aquat. Bot., 47(3-4), 197-212.
Tang, K. W., D. F. McGinnis, D. Ionescu, and H. P. Grossart (2016), Methane Production in Oxic Lake Waters
Potentially Increases Aquatic Methane Flux to Air, Environ. Sci. Technol. Lett., 3(6), 227-233.
Yao, M. Y., C. Henny, and J. A. Maresca (2016), Freshwater Bacteria Release Methane as a By-Product of
Phosphorus Acquisition, Appl. Environ. Microb., 82(23), 6994-7003.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Analysis of the layer structure of thermal microstructure profiles
of stratified lakes: new insights into vertical fluxes?
A.M. Folkard1*
1
Lancaster Environment Centre,
Lancaster University, Lancaster, United Kingdom
*Corresponding author, e-mail a.folkard@lancaster.ac.uk
KEYWORDS
Lakes; stratification; mixing; thermal microstructure; turbulence
EXTENDED ABSTRACT
Introduction
Thermal microstructure profiling has been an established technique for investigating
mixing and stratification in lakes and oceans for many years (e.g. Wuest and Lorke, 2003).
Most commonly, data from these instruments are analysed in the frequency domain to extract
measures of vertical turbulent diffusivity (K Z). These may then be used to quantify the rate of
vertical turbulent mixing in models of lake heat budgets, nutrient cycling, plankton population
dynamics etc. Here, I explore a different approach to these data, focussing on their spatial
domain characteristics. In particular, I analyse the multi-scale layered structure of their
temperature profiles. I also propose a novel, simple approach to segmentation of the profiles
into self-contained mixing layers, and compare values of Thorpe-scaled-derived turbulent
diffusivity with those derived from classical curve-fitting methods in the frequency domain.
Depth (z)
Materials and methods
The analysed profiles were all recorded with a SCAMP (Self-Contained Automatic
Micro-Profiler) (PME Inc., San Diego, USA) in Esthwaite Water, NW England (surface area
1km2; maximum depth ~15.5m). Raw profiles were pre-processed by (i) trimming at top and
bottom; (ii) interpolating to 1mm resolution; and (iii) removing noise, following Gargett and
Garner (2008) using a temperature threshold of 3.5u10-3qC.
Segmentation was carried out by identifying “self-contained” depth sections of the profile
i.e. sections where all the data points in the Thorpe-ordered profile also appeared in the
original profile, implying no mixing between different sections due to overturns in the profile.
Thorpe-scale-derived turbulent diffusivity was calculated as KZ = 1.6Q1/2LTN1/2 (Shih et
al., 2005) for every point in the profile (Thorpe scale, LT, and buoyancy frequency, N, being
calculated using centred 0.5m windows). KZ was also calculated using the Batchelor curvefitting method (Luketina and Imberger, 2001), and the results of the two methods compared.
The multi-scale layering of the
T profile
dT/dz
d2T/dz2
(Thorpe-ordered) profiles was carried out
mixed layer
using “rulers” (straight best fit lines) of
lengths from 3mm to the full profile length.
“trough”
“shoulder”
The gradient (dT/dz) of the ruler centred at
stratified layer
each point in the profile was recorded,
“shoulder”
“peak”
mixed layer
giving a full dT/dz profile for each ruler
length. Second derivative (d2T/dz2) profiles
Figure 1: Layer-identification process
were then calculated, using the same spatial
(repeated at multiple scales)
scale in each case.
At depths where d2T/dz2 peaks, the temperature profile is changing most rapidly from a
low-gradient (well-mixed) section to a high-gradient (highly-stratified) section; where there
2
Physical Processes in Natural Waters 2015
are troughs in d2T/dz2, the temperature profile is changing most rapidly the other way. Thus,
these points identify “shoulders” in the temperature profile that distinguish relatively-mixed
layers from relatively-stratified ones (Figure 1). The total number of these shoulders was
calculated for each ruler-length, and a pseudo-spectrum (ruler length vs. number of
shoulders/layers) then plotted for each main section (active mixing layer; mixed layer;
metalimnion, hypolimnion, benthic boundary layer) of each profile.
Number of layers (L)
Depth (m)
Results and discussion
Typically, the segmentation identified - as a single segment - a relatively large section at
the top of each profile. This is the actively-mixing surface layer, and is distinguished from the
surface mixed layer (Gregg and Brainerd,
0
1995). Beneath this, the segments are
-2
typically very small (<10mm), and
interspersed with occasional overturns
-4
(Figure 2).
-6
The Thorpe-scale-derived KZ was found
-8
to fall within the range of variation of the
Batchelor curve-fitting KZ in the strongly-10
stratified metalimnion, but to underestimate
-12
it in less-stratified regions.
The pseudo-spectra of the layered
-14
8
10
12
14
16
structure of the profile have an
Temperature (oC)
approximately fractal form, in that they fit
Figure 2: Example temperature profile
well a constant power law relationship
with segmentation
between the ruler scale and the number of
layers, where the index of the power law, D, is identified as the fractal dimension (Figure 3).
10000
The fractal dimensions of the profile sections
L = 22765*R-1.37
vary consistently with respect to N, LT and KZ.
D = 1.37
Deviations of the pseudo-spectra from perfect
1000
fractality (i.e. deviations of the plotted pseudospectra from best-fit power law lines) are
100
observed and interpreted as indicating moreand less-commonly occurring spatial scales
10
within the layering structure, and therefore
preferential mixing scales and differing
degrees of completeness of mixing. This
1
1
10
100
1000
10000
suggests that estimates of vertical mixing rates
Ruler Length (R, mm)
in lakes should consider the effects of
Figure 3: Example of a profile pseudo-spectrum,
differences in small-scale stratification
showing
increase in number of layers (L) in profile
structure, as they may indicate barriers to (or
with decrease in resolution scale (ruler length, R) and
facilitations of) vertical fluxes not picked up
best fit power law line with fractal dimension D
by larger-scale estimates of diffusivity.
REFERENCES
Gargett, A. and T. Garner (2008) Determining Thorpe scales from ship-lowered CTD density profiles, J. Atmos.
Ocean. Tech., 25, 1657-1670.
Gregg, M.C. and K.E. Brainerd (1995) Surface mixed and mixing layer depths, Deep-Sea Res. I, 42, 1521-1543.
Luketina, D.A. and J. Imberger (2001) Determining Turbulent Kinetic Energy Dissipation from Batchelor Curve
Fitting, J. Atmos. Ocean. Tech., 18, 100-113.
Shih, L. H., J.R. Koseff, G.N. Ivey, and J.H. Ferziger (2005) Parameterization of turbulent fluxes and scales
using homogeneous, sheared, stably-stratified turbulence simulations, J. Fluid Mech., 525, 193-214.
Wuest, A., and A., Lorke (2003) Small-scale hydrodynamics in lakes, Ann. Rev. Fluid Mech. 35, 373-412.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Lake-atmosphere exchange of CO2 and CH4 in arctic
Siberia
D. Franz1,8*, J. Boike2, G. Kirillin3, Ivan Mammarella4, Timo Vesala4,5, N. Bornemann2, E.
Larmanou1,6, M. Langer2,7, and T. Sachs1
1
Department of Geodesy, GFZ German Research Centre for Geosciences, Potsdam,
Germany
2
Alfred Wegener Institute for Polar and Marine Research, Potsdam, Germany
3
Leibniz-Institute of Freshwater Ecology and Inland Fisheries (IGB), Department of
Ecohydrology, Berlin, Germany
4
Department of Physics, University of Helsinki, Helsinki, Finland
5
Department of Forest Sciences, University of Helsinki, Helsinki, Finland
6
now at: Swedish University of Agricultural Sciences, Department of Forest Ecology and
Management, Umeå, Sweden
7
now at: Humboldt-Universität zu Berlin, Department of Geography, Berlin, Germany
8
now at: Thünen Institute of Climate-Smart Agriculture, Braunschweig, Germany
*Corresponding author, e-mail daniela.franz@thuenen.de
KEYWORDS
Arctic; thermokarst lake; CO2 and CH4 flux, eddy covariance, spring peak.
EXTENDED ABSTRACT
Introduction
Increased carbon release due to permafrost thaw and degradation is considered as
intensive positive feedback mechanism of climate change (Schuur et al. 2015). Thermokarst
lakes, representing a typical component of circum-Arctic permafrost landscapes, are effective
processors of organic carbon and emitters of CH₄, especially in Yedoma permafrost which is
rich in ice and organic carbon (Walter et al. 2006). However, in general, arctic lakes are
particularly underrepresented in observational studies on lake-atmosphere greenhouse gas
(GHG) exchange, as they are most often remote and challenging measurement sites. This is
particularly true for the period of ice break-up, which is assumed to result in a spring emission
peak especially of CH4 that could contribute considerably to the annual GHG emissions. In
order to prove the existence of such an emission burst and increase the knowledge on arctic
lake-atmosphere exchange we applied eddy covariance (EC) flux measurements on a Siberian
thermokarst lake. Our specific research questions are: How relevant is the spring ice break-up
with regard to the greenhouse gas (GHG) balance of the Siberian thermokarst lake? How do
the carbon fluxes differ between ice-covered and open-water conditions?
Materials and methods
The studied lake is located in the Yedoma landscape of Kurungnakh Island in the
southern-central part of the Lena River Delta (72° 17.9’ N, 126° 10.4’ E), which belongs to
the zone of continuous permafrost. As common for the Arctic, lakes and ponds capture a
considerable proportion of land surface in the delta. The thermokarst origin of lakes is
common in the lowland tundra permafrost areas of Northeast Siberia, which have a high to
Physical Processes in Natural Waters 2015
2
moderate ground ice content and thick sediment cover, similar to the arctic lowlands in central
and eastern Siberia, interior and northern Alaska and northwest Canada (Boike et al. 2015).
The studied lake has an area of approximately 1.25 km and a mean depth of about 3.1 m. The
lake is ice-covered for 8-9 months each year and characterized as floating-ice (Arp et al.
2012) and polymictic lake with only a few days of summer stratification (Boike et al. 2015).
In April 2014 we positioned a floatable measurement platform on the 1.7 m thick icecover in the centre of the lake. We equipped the platform with eddy covariance
instrumentation and measured turbulent fluxes of momentum, heat, H2O, CO2 and CH4. We
chose the LICOR enclosed- (LI7200) and open-path analysers (LI-7700) due to the limitations
of power, which was supplied by four solar panels. Furthermore, data on basic atmospheric
variables (air temperature, relative humidity, radiation), snow depth, water temperatures and
platform position as well as time lapse pictures were collected onboard the raft. Continuous
EC flux and additional measurements cover four months including late winter ‘frozen’ icecover conditions, the ice-cover melt (starting 21 May 2014) and ice break-up, as well as about
two-thirds of the annual open-water season (24 June until mid of August 2014). Half-hourly
fluxes were computed with EddyPro 6.1.0 (LI-COR, Nebraska, US) following common
procedures.
Results and discussion
Our measurements do not support the hypothesis of a spring CH4 emission burst
associated with ice break-up. Instead, the break-up and associated full water-column mixing
yielded an immediate shift in CH4 and CO2 exchange behaviour of the lake. The ice-cover
period was characterized by small net CH4 emission with occasional peaks and very variable
CO2 flux rates around zero with a tendency towards emission and uptake during ‘frozen’ and
‘melting’ conditions, respectively. The daytime uptake of CO2 with increasing temperatures in
spring was already observed for sea ice (Heinesch et al. 2009, see also Sørensen et al. 2014),
and related to brine volume and associated brine pCO2 (surface partial pressure of CO2).
Wavelet coherence analysis indicated correlations between CO2 flux and heat fluxes during
ice-covered conditions. For sea ice Sørensen et al. (2014) observed heat fluxes to be aligned
with the surface resistance controlling the vertical transport of CO2 between the atmosphere
and the surface. In comparison to ice-covered conditions, the lake was a clear source of CO2
and CH4 during the open-water period.
We expect three processes causing the missing CH4 burst: a) consistent emissions from
already ice-free areas (e.g. close to the shore) during melting, indicated by occasional higher
emission rates during that period, b) the progressive release of CH4 trapped in frozen bubbles
during ice-cover melt (see e.g. Walter et al. 2006), and c) continuous methane oxidation
below the ice-cover (see Denfeld et al. 2016). Supporting the latter, we recognized clear and
slightly increased CO2 emissions for the first ice-free week, when the water column was
mixed completely for the first time after winter. During this time CO2 and CH4 release
showed a similar temporal pattern (wave length > 1 day), especially pronounced in case of
CO2.
Wavelet analysis and coherence illustrated changes of CO2 flux behaviour during a threeweeks period in summer, when the drifting platform was located close to the eastern shore of
the lake. The diurnal cycle was more pronounced (with nighttime peak CO2 emissions and
daytime uptake) than during times when the platform was further away from the shore. CO2
correlated with short-wave incoming radiation on a daily scale indicating photosynthesis and
thus the inclusion of the tundra within the EC source area. During this period we further
observed stronger CH4 emissions throughout the day, potentially highlighting the impact of
shallow water areas in the EC footprint. Thermokarst lakes are characterised by continuous
erosion of the lake shores and thus refill of organic material into the shallow lake areas during
Franz et al.
3
the open-water period. As soon as the raft was brought back closer to the centre of the lake,
the flux dynamics switched back to the ‘lake flux behaviour’, i.e. clear net release of CO 2 and
lower CH4 release rates, indicating in sum a clear source of carbon GHGs. However, CH4
release in August was much stronger than in June when the lake became ice-free, probably
due to higher water and sediment temperatures.
REFERENCES
Arp, C. D., Jones, B. M., Liljedahl, A. K., Hinkel, K. M. and Welker, J. A. (2015), Depth, ice thickness, and iceout timing cause divergent hydrologic responses among Arctic lakes, Water Resour. Res., 51, 9379-9401.
Boike, J., Georgi, C., Kirilin, G., Muster, S., Abramova, K., Fedorova, I., Cetverova, A., Grigoriev, M.,
Bornemann, N. and Langer, M. (2015), Thermal processes of thermokarst lakes in the continuous
permafrost zone of northern Siberia - Observations and modeling (Lena River Delta, Siberia),
Biogeosciences, 12, 5941-5965.
Denfeld, B. A., Ricão Canelhas, M., Weyhenmeyer, G. A., Bertilsson, S., Eiler, A. and Bastviken, D. (2016),
Constraints on methane oxidation in ice-covered boreal lakes, J. Geophys. Res. Biogeosci., 121, 19241933.
Heinesch, B., Aubinet, M., Carnat, G., Geilfus, N.-X., Goossens, Eicken, H., Papakyriakou, T., Petrich, C.,
Tison, J.-L., Yernaux, M. and Delille, B. (2009), Survey of air-ice-ocean carbon dioxide exchanges over
Arctic sea ice. 8th International Carbon Dioxide Conference September 13-19, Jena, Germany, extended
abstract.
Morgenstern, A. (2012), Thermokarst and thermal erosion: Degradation of Siberian ice-rich permafrost, PhD
thesis, Universität Potsdam.
Schuur, E. A. G., McGuire, A. D., Schädel, C., 2015, Climate change and the permafrost carbon feedback,
Nature, 520, 171-179, 2015.
Sørensen, L. L., Jensen, B., Glud, R. N., McGinnis, D. F., Sejr, M. K., Sievers, J., Søgaard, D.H., Tison, J.-L.
and Rysgaard, S. (2014), Parameterization of atmosphere–surface exchange of CO2 over sea ice,
Cryosphere, 2014, 8(3), 853-866.
Walter, K. M., Zimov, S. A., Chanton, J. P., Verbyla, D. and Chapin III, F. S. (2006), Methane bubbling from
Siberian thaw lakes as a positive feedback to climate warming, Nature, 443, 71-75.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Harp Lake model intercomparison experiment: focus on vertical
gas transfer
S. Guseva 1, V. Stepanenko 1,2, W. Thiery 3, Z. Tan 4, Q. Zhuang 5,
K. Jöhnk 6, B.A. Polli 7, T. Bleninger 8, W. Wang 9, H. Yao 10
1
Department of Geography, Lomonosov Moscow State University ,
Moscow, Russia
2
Laboratory for Supercomputing modelling of climate system processes,
Moscow State University, Moscow, Russia
3
Department of Earth and Environmental Sciences, University of Leuven,
Leuven, Belgium
4
Pacific Northwest National Laboratory, Richland, Washington, USA
5
Department of Earth, Atmospheric, and Planetary Sciences, Purdue University,
West Lafayette, Indiana, USA
6
CSIRO Land and Water, Black Mountain, GPO Box 1666,
Canberra ACT 2601, Australia
7
Federal University of Paraná (UFPR), Curitiba, Paraná, Brazil
8
Graduate Program on Water Resources and Environmental Engineering,
Federal University of Paraná, Brazil
9
Faculty of Biology and the Environment, Nanjing Forestry University,
Nanjing, China
10
Dorset Environmental Science Centre, Ontario Ministry of Environment, Dorset,
Ontario, P0A 1E0 Canada
*Corresponding author, e-mail guseva.sofya.pavlovna@gmail.com
KEYWORDS
LakeMIP; numerical simulation; lakes; gas transfer; diffusivity.
EXTENDED ABSTRACT
Introduction
Over the last decade the field of lake modeling has attracted much attention in weather
and climate modelling community. Several lake models of diverse complexity have been
developed and incorporated in the numerical weather forecast systems and climate models.
Hence, the comparison of lake models and their verification are essentially relevant.
Especially it applies to simulation of biogeochemical processes and gas transfer in lakes. One
of the projects, accumulating knowledge on the lake models’ “behavior” is LakeMIP (Lake
Model Intercomparison Project), launched in 2008 [Perroud et al., 2009; Stepanenko et al.,
2010; Thiery et al., 2014, etc.]. Previous studies investigated the ability of the different lake
models to represent the thermodynamic state of the water bodies in different latitudes. This
research focuses on the simulation of key factors of greenhouse gas dynamics in lakes, such
as the lake stratification, the diffusion of gases, and the ice cover.
Materials and methods
The study site is a large (71.4 ha), deep (hmax = 37.5 m; hmean = 13.32 m) oligotrophic
Harp Lake, located in south-central Ontario, Canada (45° 23’ N, 79° 07’ W). It is enclosed by
Physical Processes in Natural Waters 2015
2
forest catchment [Cox, 1978; Dillon et al., 1978]. Harp Lake has six inflows and one outflow
[Yao et al., 2014]. The observation dataset contains the time series (14.07.2010 – 19.10.2015)
of meteorological variables such as: temperature, pressure, precipitation, wind speed,
shortwave and longwave downward radiation. In addition, it includes the evolution of
temperature profile (up to 27.1 m), oxygen profile (up to 35 m, without winter time period,
time step – 1 day) , oxygen (at depths 1 m, 18 m, time step – 1 hour) and carbon dioxide (at a
depth 0.39 m, time period: 12.03.2012 – 19.10.2015). The meteorological forcing is used as
an identical input for all models.
Among the participating models there are those using Henderson-Sellers-based
diffusivity [Henderson-Sellers, 1985] such as: bLake4Me [Tan et al., 2015], FAQ-DNDC
[Wang et al., 2016], MTCR-1 [Polli and Bleninger, 2015] ; and k-ε turbulence closure models
- LAKE [Stepanenko et al., 2011; Stepanenko et al., 2016], LAKEoneD [Jöhnk and Umlauf,
2001 ; Jöhnk et al., 2008] and the lake model FLake employing a concept of self-similarity of
the temperature profile [Mironov, 2003; Mironov, 2008]. The set of numerical experiments
includes a baseline experiment and other experiments testing the sensitivity of models to
variation of the extinction coefficient, the lake depth, and the initial conditions. Additionally,
in order to understand the role of eddy diffusivity for the gases, an experiment solving the
vertical transport of passive tracer governed by eddy diffusivities from lake models was
conducted.
Results and discussion
Comparing the simulation results allows to select features of models, affecting the
vertical distribution of gases in the lake.
The thermodynamic state of Harp Lake (Fig.1), its seasonal variation, sensitivity to the
extinction coefficient was adequately represented by all models described above, excepting
the Flake model. It produces (as in previous LakeMIP studies) an underestimated temperature
gradient in the thermocline, extended to the deep layers, not revealed in observation data. It
can be a source of errors in reproducing the distribution of gases by this model in future.
The simulation of the ice-cover period was challenging for all models. Only bLake4Me,
FAQ-DNDC, LAKE with more sophisticated approaches of the ice formation and degradation
could perform satisfactorily in simulating the onset of ice-cover and ice-off dates.
It has been suggested that temperature in water column largely depend on turbulent
diffusivity [Stepanenko et al., 2014]. It possibly also has a crucial role for the transfer of
gases. In order to identify “clear” effect (without any biogeochemical influence) of the
diffusivity on the vertical distribution of a passive tracer the diffusion equation has been
solved considering the vertical turbulent coefficient from the models’ output. The results of
the modelling (bLake4Me, MTCR-1, LAKEoneD, LAKE) demonstrate (Fig.2), that the one
order difference of the eddy diffusivity can impact the distribution of substance in a lake. The
events of the autumn and spring emissions connected to seasonal overturning are very
sensitive to the details of a turbulence parameterization. Reducing the depth of lake
approximately by half, the mean integral concentration of the tracer in water column
decreases from 24 % to 76 %, in different models. Meanwhile, the mean flux, calculated as:
Fsurf
Fbottom
§ C Ceq ·
K ¨ w
¸ / Fbottom , where Fsurf – the flux at the air-water interface; Fbottom –
z
'
©
¹
the constant flux at the bottom of a lake; K – the eddy diffusivity coefficient from lake models,
Ceq - the concentration at the surface of a lake ; Cw – the concentration in the underlying water,
'z - the thickness of the finite difference grid layer –
Guseva et al.
3
increases from 4 % to 54 %, primarily because of eddy diffusivity variability during
summer stratification.
The most sophisticated lake models with biogeochemical modules such as bLake4Me and
LAKE demonstrate well representation of the production and dynamics of oxygen. The mean
concentration of carbon dioxide in model bLake4Me is greater than in LAKE, probably due to
involving the transport from catchment by inflows and less intensive vertical mixing.
However, in both models correlation to the observation data ( CCO2 ) is small: 0.2-0.3.
Physical Processes in Natural Waters 2015
4
Fig.1. Time-depth profiles of temperature in Harp Lake, reference model run and
observation data. The grey boxes indicate duration of the ice-cover period, the white patterns a lack of data
Fig.2. Time-depth profiles of a passive tracer in a lake, reference model run
REFERENCES
Cox, E.T. (1978). Counts and measurements of Ontario lakes: watershed unit summaries based on maps of
various scales by watershed unit. Ontario Ministry of Natural Resources.
Dillon, P. J., R. A. Reid, and E. De Grosbois (1987). The rate of acidification of aquatic ecosystems in Ontario,
Canada. Nature, 329, 45-48.
Henderson-Sellers, B. (1986). New formulation of eddy diffusion thermocline models. Applied Mathematical
Modelling, 9(6), 441-446.
Jöhnk, K.D. and L. Umlauf (2001) . Modelling the metalimnetic oxygen minimum in a medium sized alpine
lake. Ecological Modelling, 136(1), 67–80.
Jöhnk, K. D., J.E.F. Huisman, J. Sharples, B.E.N. Sommeijer, P. M. Visser, and J. M. Stroom (2008). Summer
heatwaves promote blooms of harmful cyanobacteria. Global change biology, 14(3), 495–512.
Mironov D. V. (2003). Parameterization of lakes in numerical weather prediction. Part 1: Description of a lake
model. Available from the author, Dmitrii. Mironov@ dwd. de.
Mironov D. V. (2008) Parameterization of lakes in numerical weather prediction.Description of a lake model.
Cosmo Technical Report 11.
Perroud, M., Goyette, S., Martynov, A., Beniston, M., & Anneville, O. (2009). Simulation of multiannual
thermal profiles in deep Lake Geneva: a comparison of one-dimensional lake models. Limnology and
Oceanography-Methods, 54(5), 1574-1594.
Guseva et al.
5
Polli B. A. and Bleninger T. (2015). Modelagem do transporte de calor no reservatÓrio Vossoroca. 21 Simposio
Brasileiro De Recursos Hidricos.
Stepanenko, V. M., Goyette S., Martynov A., Mironov D., (2010). The LakeMIP – Lake Models
Intercomparison Project. Boreal Env. Res., 15, 191-202.
Stepanenko, V.M., Machul’skaya E.E., Glagolev M.V., Lykossov V.N. (2011). Numerical modeling of methane
emissions from lakes in the permafrost zone. Izvestiya, Atmospheric and Oceanic Physics, 47(2), 252–264.
Stepanenko, V., Jöhnk, K. D., Machulskaya, E., Perroud, M., Subin, Z., Nordbo, A., ... & Mironov, D. (2014).
Simulation of surface energy fluxes and stratification of a small boreal lake by a set of one-dimensional
models. Tellus A, 66.
Stepanenko V., Mammarella I., Ojala A., Miettinen H., Lykosov V., and Vesala T. (2016) . Lake 2.0: a model for
temperature, methane, carbon dioxide and oxygen dynamics in lakes. Geoscientific Model Development,
9(5), 1977–2006. doi: 10.5194/gmd-9-1977-2016. URL http://www. geosci-model-dev.net/9/1977/2016/.
Tan Z., Zhuang Q., and Anthony K. W. (2015) Modeling methane emissions from arctic lakes: Model
development and site-level study. Journal of Advances in Modeling Earth Systems, 7(2),459–483.
Thiery, W., Stepanenko V., Fang X., Jöhnk K., Li Z., Martynov A., Perroud M., Subin Z., Darchambeau F.,
Mironov D., and N. van Lipzig. (2014). LakeMIP Kivu: evaluating the representation of a large, deep
tropical
lake
by
a
set
of
one-dimensional
lake
models.
Tellus
A,
66.
doi:http://dx.doi.org/10.3402/tellusa.v66.21390.
Wang, W., Roulet, N. T., Strachan, I. B., and Tremblay, A. (2016). Modeling surface energy fluxes and thermal
dynamics of a seasonally ice-covered hydroelectric reservoir. Science of the Total Environment, 550, 793805.
Yao, H., Samal, N. R., Jöhnk, K. D., Fang, X., Bruce, L. C., Pierson, D. C., ... and James, A. (2014). Comparing
ice and temperature simulations by four dynamic lake models in Harp Lake: past performance and future
predictions. Hydrological Processes, 28(16), 4587-4601.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Evaluation of methane emission from a mid-latitude lake with the
eddy covariance technique
H. Iwata1*, R. Hirata2, Y. Takahashi2, M. Ito3, K. Iizuka3, Y. Miyabara1, D. Kobayashi1, and
T. Tokida4
1
Department of Environmental Sciences, Shinshu University, Matsumoto, Japan
Center for Global Environmental Research, National Institute for Environmental Studies,
Tsukuba, Japan
3
Center for Southeast Asian Studies, Kyoto University, Kyoto, Japan
4
Division of Biogeochemical Cycles, National Institute for Agro-Environmental Sciences,
Tsukuba, Japan
2
*Corresponding author, e-mail hiwata@shinshu-u.ac.jp
KEYWORDS
Dissolved CH4 concentration; ebullition; flux partitioning; stable isotope; wavelet analysis.
EXTENDED ABSTRACT
Introduction
Lake ecosystems are one of the major methane (CH 4) sources. CH4 emission from lakes
to the atmosphere occurs as diffusion through the water column, ebullition, and plantmediated transport. It was reported that CH4 emission via ebullition contributed up to 80 % of
the total (Bastviken et al., 2004), thus it can be the major emission pathway in some lake.
However, such evaluations have been based on discontinuous observations such as the
floating chamber technique, and it may have resulted in an underestimation of the ebullition
emission (Podgrajsek et al., 2014a) due to the sporadic nature of ebullition both in time and
space. For an accurate estimation of CH4 emission, the eddy covariance technique has an
advantage in that it can cover footprint of 104 m2. The application of the eddy covariance
technique for CH4 emission in lakes so far has been limited to a few studies (e.g., Eugster et
al., 2011; Podgrajsek et al., 2014b). Thus, more eddy covariance studies are needed to
understand the detail of environmental controls on CH 4 emission from lakes.
This study reports CH4 emission from a shallow lake in Japan observed with the eddy
covariance technique along with periodic measurements of dissolved CH4 concentration.
Materials and methods
Observations were conducted in Lake Suwa, a eutrophic lake, which has an area of 12.8
2
km and maximum depth of 7.6 m. Eddy covariance instruments were installed on a mast at a
pier located on the southeast coast. The instruments include a three-dimensional ultrasonic
anemothermometer (CSAT3, Campbell Scientific, USA) and an open-path CH4 analyzer (LI7700, Li-Cor, USA). Relevant observations of atmospheric and lake environment were also
conducted. The observation started in June, 2016 and still ongoing. Lake water was sampled
approximately once a month to analyse for dissolved CH 4 concentration profile. Flux data
used in this study were corrected for the high-frequency loss, air density fluctuation, and
spectroscopic effect. Only data with wind direction from the lake were used.
Results and discussion
Physical Processes in Natural Waters 2015
2
In Lake Suwa large CH4 emission occurred in specific wind directions, with the
maximum emission up to approximately 2.0 μmol m-2 s-1 in July. We confirmed that steady
ebullition occurred in this direction. The trapped bubble was analysed for the CH4
concentration and stable isotope ratio, and we found the concentration was 91 % and δ13CCH4 was -62.8 ‰. The δ13C-CH4 value suggests that the emitted CH4 through the steady
ebullition had the origin of microbial decomposition.
When data with wind direction from the steady ebullition area were excluded from the
analysis, CH4 emission had a clear seasonal variation, typically with 0.3 μmol m-2 s-1 in
summer and 0.1 μmol m-2 s-1 in winter, similar to a variation in the temperature near the lake
bottom. However, still large CH4 emissions up to approximately 1.0 μmol m-2 s-1 were
observed sporadically in these directions.
Dissolved CH4 concentration showed the highest concentration in July, with higher
concentration in deeper layer. The maximum concentration was 8.7 μmol L-1 near the lake
bottom. Subsequently, dissolved CH4 concentration decreased, but still higher concentration
in deeper layer. After October, concentration was nearly constant through the profile, which
was the results of complete lake water mixing.
In analysing the response of CH4 emission to environmental variables, it is desirable to
partition the net flux into emissions through diffusion and ebullition processes. Here, we have
developed such a flux partitioning technique for CH 4 flux. The 10 Hz turbulence data showed
that the scalar similarity between air temperature (or water vapor concentration) and CH4
concentration held when CH4 emission was low, whereas the similarity did not hold when
CH4 emission was high due to abrupt large positive deviations of CH4 signal from the time
average. Ebullition occurs sporadically both in time and space and emits bubbles of high CH4
concentration, thus it can lead to such abrupt large positive deviations, resulting in the scalar
dissimilarity. We hypothesized that similar fluctuation components reflect diffusion emission
and dissimilar fluctuation components reflect ebullition emission, and the actual turbulent
fluctuations are the superposition of both. By separating the similar and dissimilar
components based on wavelet coefficients, we partitioned CH4 flux into emission due to
diffusion and ebullition. We found that diffusion emission was controlled by wind speed on a
short time scale, rather than temperature near the lake bottom. This suggests that transport
efficiency in the water column, rather than CH4 production in the sediment, has stronger
influence on the diffusion emission in this lake.
In summary, the one-year observation of CH4 flux revealed that both steady and sporadic
ebullition emission occurred in Lake Suwa. When these ebullition emissions were excluded,
the seasonal variation in CH4 emission was controlled by CH4 production in the lake sediment
layer. On the shorter time scale, CH4 emission was influenced by the transport efficiency in
the water column. The proposed flux partitioning technique needs further validation, but it
can readily be applied to other data sets, since it only requires high-frequency turbulence data
with a few empirical parameters.
REFERENCES
Bastviken, D., J. Cole, M. Pace, and L. Tranvik (2004), Methane emissions from lakes: dependence of lake
characteristics, two regional assessments, and a global estimate, Global Biogeochemical Cycles, 18(4),
GB4009.
Eugster, W., T. DelSontro, and S. Sobek (2011), Eddy covariance flux measurements confirm extreme CH4
emissions from a Swiss hydropower reservoir and resolve their short-term variability, Biogeosciences, 8,
2815-2831.
Podgrajsek, E., E. Sahlée, D. Bastviken, J. Holst, A. Lindroth, L. Tranvik, and A. Rutgersson (2014a),
Comparison of floating chamber and eddy covariance measurements of lake greenhouse gas fluxes,
Biogeosciences, 11, 4225-4233.
Podgrajsek, E., E. Sahlée, and A. Rutgersson (2014b), Diurnal cycle of lake methane flux, Journal of
Geophysical Research: Biogeosciences, 119(3), 236-248.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Methane oxidation in Lake Kuivajärvi
T. Saarela1, A.J. Rissanen2, H. Jäntti1*, S. Aalto 3, M. Tiirola3, J. Pumpanen1 and A. Ojala4
1
Department of Environmental and Biological Sciences,
University of Eastern Finland, Kuopio, Finland
2
Laboratory of Chemistry and Bioengineering,
Tampere University of Technology, Tampere, Finland
3
Department of Biological and Environmental Sciences,
University of Jyväskylä, Jyväskylä Finland
4
Department of Environmental Sciences,
University of Helsinki, Helsinki, Finland
*Corresponding author, e-mail helena.jantti@uef.fi
KEYWORDS
Boreal lakes; methane; methane oxidation; anoxia.
EXTENDED ABSTRACT
Introduction
Atmospheric methane (CH4) is a global greenhouse gas that is emitted to the atmosphere
by both natural and anthropogenic sources (Dlugokencky et al., 2011). One of most
significant natural sources of CH4 in the boreal region are lakes that mineralize carbon
produced within the lake and its catchment. Majority of the CH4 in lakes is formed in the
anoxic lake sediment layers, although CH4 can also enter lakes from the catchment by surface
runoff (Miettinen et al., 2015). While sediments produce substantial amounts of CH4, very
small portion of it enters the water column because CH4 is oxidized to carbon dioxide (CO2)
in the oxic sediment surface. However, the oxic layer can disappear in late summer when
mineralization of organic matter consumes all oxygen from the sediment surface.
Consequently, CH4 is released from the sediment to the water column. CH4 oxidation can also
take place in the water column (Kankaala et al., 2006), but the rates and the rate controlling
factors are still poorly known. To gain better understanding of the effect of bottom water
anoxia on the CH4 emissions from boreal lakes, we measured the CH 4 and CO2 gas
concentrations and their į13C values from seasonally anoxic Lake Kuivajärvi near Hyytiälä
Forest Research Station during open water season 2016. In addition, we measured the
potential CH4 oxidation rates in the water column after it had changed anoxic.
Materials and methods
Water sampling was done 4 times between May and September in 2016 at the deepest
point of the lake (~12 m) (Table 1). In addition to nutrients and CH4 and CO2 gas
concentrations, we also measured the stable isotopic signatures of CH4 and dissolved
inorganic carbon (DIC). The stable isotopic signature of CH4 į13C) can be used as an
indicator for CH4 production and consumption because newly formed CH4 has a distinctly
light isotopic signature (į13C = -100 -45 ‰) (Grey 2016). When CH4 is oxidized, 12C is
preferred over 13C, hence the į13C- value of the remaining CH4 increases. Because the į13Cvalue does not indicate how fast CH4 is oxidized, we also measured the potential CH4
oxidation rate by adding 13C-labelled CH4 (99% 13C-label) to the samples and following the
transfer of 13C label from CH4 to DIC pool.
Physical Processes in Natural Waters 2017
2
Table 1. The sampling dates and parameters analyzed in each sampling time.
Date
Analyses
25.5.2016 T, pH, O2, CH4, CO2
18.7.2016 T, pH, O2, NO3-, NH4+, SO42-, CH4, CO2, į13CH4, į13DIC
15.8.2016 T, pH, O2, NO3-, NH4+, SO42-, CH4, CO2, į13CH4, į13DIC, CH4 oxidation rate
5.9.2016 T, pH, O2, NO3-, NH4+, SO42-, DOC, Fe, CH4, CO2, į13CH4, į13DIC, CH4
oxidation rate
Results and discussion
The stratification and bottom anoxia had developed by September and the CH4
concentrations in the hypolimnion peaked simultaneously, indicating that during anoxia the
CH4 was not oxidized in the sediment but instead released to the water column (Figure 1a,
1c). The changes in į13C-CH4 also confirmed the transition of CH4 oxidation zone from the
sediment to the deep water column (Figure 1b).
Figure 1. The CH4 and CO2 concentrations (a), į13CH4, į13DIC (b), O2 concentration, and
methane oxidation rates (c) in Lake Kuivajärvi in September 2016.
The potential CH4 oxidation rates remained below detection limit until September when
strong anoxia had developed in the hypolimnion (Figure 1c). Interestingly, the highest
potential rates were measured right above the sediment where the O2 concentration was the
lowest (Figure 1c). Since anaerobic CH4 oxidation taking place in Lake Kuivajärvi seems
unlikely, this phenomenon could be explained by temporal micro-oxic zones allowing aerobic
CH4 oxidation in otherwise anoxic environment.
We estimated, based on the changes in the stable isotopic signature of CH4, that in
September, approximately 60 % of produced CH4 was oxidized in the water column and 40%
entered the surface water layers. Even though lakes act as a source of CH4 especially during
the hypolimnetic anoxia, methane-oxidizing bacteria can still significantly reduce CH4
emissions from lakes to the atmosphere.
REFERENCES
Dlugokencky, E.J., Nisbet, E.G., Fischer, R. and Lowry, D (2011), Global atmospheric methane: budget,
changes and dangers. Phil. Trans. R. Soc. A., 369, 2058-2072.
Grey, J (2016), The incredible lightness of being methane-fuelled: stable isotopes reveal alternative energy
pathways in aquatic ecosystems and beyond, Front. Ecol. Evol., 4(8), 1-14.
Kankaala P., Huotari J., Peltomaa E., Saloranta T. and Ojala A (2006) Methanotrophic activity in relation to
methane efflux and total heterotrophic bacterial production in a stratified, humic, boreal lake Limnol
Oceanogr 51(2): 1195-1204.
Miettinen, H., Pumpanen, J., Heiskanen, J.J., Aaltonen, H., Mammarella, I., Ojala, A., Levula, J. and Rantakari,
M (2015), Towards a more comprehensive understanding of lacustrine greenhouse gas dynamics-two-year
measurements of concentrations and fluxes of CO2, CH4 and N2O in a typical boreal lake surrounded by
managed forests. Bor. Environ. Res. 20(1), 75-89.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Chemical enhancement of CO2 at the air-water interface
in eutrophic lakes of Quebec
Jihyeon Kim and Yves T. Prairie*
Départment des sciences biologiques,
Université du Québec à Montréal, Montréal, Canada
*Corresponding author, e-mail prairie.yves@uqam.ca
KEYWORDS
Air-water gas exchange; Chemical enhancement; Lakes; Gas exchange coefficient; CO2 hydration.
EXTENDED ABSTRACT
Introduction
A precise knowledge of gas exchange processes across the air-water interface of lakes is
necessary for properly assessing carbon budgets at regional and global scales. In contrast with
chemically non-reactive gases such as methane (CH 4) and nitrous oxide (N2O), diffusive
carbon dioxide (CO2) exchange can be enhanced by chemical reactions of CO 2 with hydroxyl
ions (OH-), especially at high pH. In eutrophic lakes where pH is relatively high, chemical
enhancement is believed to be an important factor in driving CO2 fluxes [Bolin, 1960].
However, there is still considerable disagreement between observed chemical enhancement
rates (α) and those predicted from theoretical models [Wanninkhof and Knox, 1996; Bade and
Cole, 2006]. To address this issue, we quantified the actual contributions of chemical
enhancement to CO2 fluxes between the air and water. In addition, we examined any
discrepancies that might exist between the observed α in the field and that predicted from the
calculation of Hoover and Berkshire (1969) to verify the general applicability of the model.
Materials and methods
We sampled 21 lakes in Quebec, Canada during the summer of 2015 and 2016, selected
to cover a wide range of chemical and physical conditions. Diffusive gas exchange of CO2
and that of CH4, which does not react with hydroxyl ions in water, were measured using
floating chambers. Then, α was obtained in two ways: 1) as the ratio of measured gas
exchange coefficients derived from CO2 (k600-CO2) and from CH4 (k600-CH4), and 2) from the
theoretical model of Hoover and Berkshire (H&B).
Results and discussion
The sampled lakes covered a broad range of trophic status, pH, water temperatures and
wind speeds, all of which are considered as important regulators for chemical enhancement.
We found that 58% of sites were undersaturated in CO2 with respect to the atmosphere,
leading to a net influx of CO2 into the lakes. In most lakes, k600-CO2 and k600-CH4 were similar
(on average 2.46 and 2.54 m d-1, respectively), suggesting that chemical enhancement of CO 2
was very low even under high pH conditions (8.5 to 9.5) (Figure 1). In addition, observed α
values were considerably lower than model predictions (Figure 2). Our results indicate that
the often-used theoretical model of H&B overestimates chemical enhancement of CO 2. Our
data instead suggest that chemical enhancement is not a very significant contributor to airwater CO2 flux in lakes of Quebec.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Figure 1. The relationship between the gas exchange
velocity (k600, m d-1) estimated from CO2 and CH4.
Green line and red line represent line of equality and
regression fit, respectively.
Figure 2. Observed chemical enhancement factor (α,
dimensionless) versus that predicted from the
theoretical model of Hoover and Berkshire (1969). If
the model fits well, the dots should be close to the
green line.
Likely reasons for the discrepancy between calculated and observed values of α are: 1)
high alkalinity of the sampling lakes, 2) errors associated with the assumption of H&B model
that pH within the surface boundary layer and 3) enhanced k600-CH4 resulting from
microbubble-mediated fluxes. H&B conducted tank experiments with distilled water when
they developed and first tested their model. We propose that these tank experiments are a poor
approximation of conditions in actual lake systems. Higher alkalinity means more bicarbonate
(HCO3-) in the water, and a lower probability of CO2 chemical reactions. The alkalinity of our
study lakes ranges from 400 to 1900 μeq/L. In addition, H&B utilized the stagnant boundary
layer model [Lewis and Whitman, 1924]. For the stagnant boundary layer model, it is assumed
that the CO2 concentration is in non-equilibrium within the layer while the concentration of
CO2 and carbon species are in thermodynamic equilibrium in the turbulent water body.
However, H&B assumed that the pH inside the boundary layer would be roughly same as that
of turbulent water body. Contrary to their assumption, if the pH of the boundary layer is lower
than that of turbulent water body resulting from atmospheric CO 2 absorption, the actual
chemical enhancement can be lower than what we might expect from the model. Also, recent
studies have shown the potential influence of microbubble-mediated CH4 fluxes [Prairie and
del Giorgio, 2013; Rosentreter et al., 2016]. While microbubble-mediated enhanced k600-CH4
would partially decrease the observed α, there is little to believe that it is pH driven. In
addition, the observed k600-CH4 is entirely consistent with wind based models suggesting
that microbubbles are not very important in these systems.
We expect these results can improve the estimation of gas exchange in eutrophic systems,
and have important consequences for the assessment of regional carbon budgets.
REFERENCES
Bade, D. L., and J. J. Cole (2006), Impact of chemically enhanced diffusion on dissolved inorganic carbon stable
isotopes in a fertilized lake, J. Geophys. Res. Ocean., 111(1), 1–10, doi:10.1029/2004JC002684.
Bolin, B. (1960), On the Exchange, Tellus, 12(3), 274–281.
Hoover, T. E., and D. C. Berkshire (1969), Effects of Hydration on Carbon Dioxide Exchange across an AirWater Interface, J. Geophys. Res., 74(2), 456–464, doi:10.1029/JB074i002p00456.
Lewis, W. K., and W. G. Whitman (1924), Absorption symposium, Ind. Eng. Chem., 16(12), 1215–1220.
Prairie, Y. T., and P. a. del Giorgio (2013), A new pathway of freshwater methane emissions and the putative
importance of microbubbles, Inl. Waters, 3(3), 311–320, doi:10.5268/IW-3.3.542.
Rosentreter, J. A., D. T. Maher, D. T. Ho, M. Call, J. G. Barr, and B. D. Eyre (2016), Spatial and temporal
variability of CO 2 and CH 4 gas transfer velocities and quantification of the CH 4 microbubble flux in
mangrove dominated estuaries, Limnol. Oceanogr., doi:10.1002/lno.10444.
Wanninkhof, R., and M. Knox (1996), Chemical enhancement of CO2 exchange in natural waters, Limnol.
Oceanogr., 41(4), 689–697, doi:10.4319/lo.1996.41.4.0689.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Modelling the effect of changes in air temperature and carbon
loading on CO2 in a boreal lake
P. Kiuru1,2*, A. Ojala3, I. Mammarella4, J. Heiskanen4, M. Kämäräinen5, T. Vesala4 and T.
Huttula1
1
Freshwater Centre, Finnish Environment Institute, Jyväskylä, Finland
2
Department of Physics, University of Jyväskylä, Jyväskylä, Finland
3
Department of Environmental Sciences, University of Helsinki, Lahti, Finland
4
Department of Physics, University of Helsinki, Helsinki, Finland
5
Climate Service Centre, Finnish Meteorological Institute, Helsinki, Finland
*Corresponding author, e-mail petri.kiuru@ymparisto.fi
KEYWORDS
Lakes; carbon dioxide; modelling; biogeochemistry; climate change.
EXTENDED ABSTRACT
Introduction
The importance of freshwater lakes in carbon cycling is pronounced in the boreal zone as
the lakes ventilate carbon originally fixed by the surrounding terrestrial system. The effect of
climate change on lake ecosystems is most evident in high latitudes. Higher air temperature
(Ta) shortens the duration of the ice-covered period and accelerates organic matter
degradation. Increase of precipitation and alteration of the seasonal distribution of stream flow
may elevate terrestrial carbon loading. Modelling is an efficient tool for estimating the effects
of climate change, but the number of mechanistic models simulating carbon in boreal lakes is
low. Recent advances in long-term high-frequency measurement of lake carbon dioxide (CO 2)
concentration and air-water flux (Mammarella et al., 2015) facilitate model development. The
purpose of our study was to estimate the possible implications of higher Ta and increased
carbon loading on lacustrine carbon cycle and to give insight on in-lake mechanisms behind
the potential impacts of climate change on CO 2 in a boreal lake.
Materials and methods
We used a one-dimensional process-based model for simulation of lake CO2 dynamics
(Kiuru et al., 2016). The model is an extension of a lake model MyLake (Saloranta and
Andersen, 2007), and it simulates lake thermodynamics, phosphorus, phytoplankton,
dissolved oxygen, and inorganic and organic carbon species. We calibrated the model for
Lake Kuivajärvi, a small, humic boreal lake, which is a constant source of CO2 to the
atmosphere in the present climate, using the comprehensive data available on carbon inflow
and the concentrations of CO2 and dissolved organic carbon (DOC) in the lake.
We studied the potential effects of climate chance induced warming on the CO 2 dynamics
of the lake between the control period 1980–2009 and the scenario period 2070–2099 using
downscaled air temperature data from three recent-generation global climate models (see
Lehtonen et al., 2016). The climate models were forced with representative concentration
pathway (RCP) scenarios representing intermediate (RCP4.5) and high change (RCP8.5) in
radiative forcing. Literature estimates were used for climate change impacts on lake inflow
volume. In addition, we compared the effects of 40 % increases in the concentrations of
inflow DOC and CO2 under the RCP4.5 scenario.
Physical Processes in Natural Waters 2015
2
Results and discussion
The near-surface CO2 concentrations were substantially higher in the scenario period
(Figure 1a, b), especially under the high forcing scenario RCP8.5. Higher wintertime inflow
accumulated more CO2-rich water in the surface layers under ice in the scenario period, but
the maximum water-column average CO2 concentration under ice was smaller and the spring
peak in CO2 air-water flux was lower because of shorter ice-covered period (Figure 1e) when
only changes in Ta and discharge volume were considered. The increase of annual CO2 flux to
the atmosphere due to higher Ta and altered seasonal distribution of inflow was 17–20 % in
the RCP4.5 scenario and 33–38 % in the RCP8.5 scenario. The projected annual mean
temperature increases of 2.8–3.6 °C and 5.3–5.8 °C resulted in shortening of the ice-covered
period by 44–49 and 69–88 days in RCP4.5 and RCP8.5, respectively.
Figure 1. Simulated CO2 concentrations (mmol m-3) in Lake Kuivajärvi in the control period 1980–2009
(dashed lines) and the scenario period 2070–2099 (solid lines) at the depths of 1 m (a, b) and 9 m (c, d) using
RCP4.5 (a, c) and RCP8.5 (b, d) as climate forcing, and CO2 air-water fluxes in the control period and without
changes in the inflow carbon concentrations (baseline) in the RCP4.5 scenario period (e), with a 40 % increase in
the inflow CO2 concentration (f), and with a 40 % increase in the inflow DOC concentration (g).
The increase in direct CO2 loading raised the simulated air-water flux of CO2 more than a
similar increase in DOC loading (Figure 1f, g). The impact of increased DOC loading on CO2
concentration was moderate because of the supposed rather refractory nature of inflow DOC
although we used a loading scenario with a substantial concentration increase. However, the
uncertainties in the climate scenarios and in the simple estimate of changes in inflow used in
this study as well as the sources of uncertainty in the biochemical model due to lack of
knowledge of lacustrine carbon system functioning need to be recognized.
REFERENCES
Kiuru, P., A. Ojala, I. Mammarella, J. Heiskanen, T. Vesala, and T. Huttula (2016), A process-based model for
simulation of lake oxygen and dissolved inorganic carbon, in Proceedings of the 19th International
Workshop on Physical Processes in Natural Waters: PPNW2016, edited by D. Wain and L. Bryant, pp. 4950, University of Bath.
Lehtonen, I., M. Kämäräinen, H. Gregow, A. Venäläinen, and H. Peltola (2016), Heavy snow loads in Finnish
forests respond regionally asymmetrically to projected climate change, Nat. Hazards Earth Syst. Sci.,
16(10), 2259-2271, doi:10.5194/nhess-16-2259-2016.
Mammarella, I., A. Nordbo, Ü. Rannik, S. Haapanala, J. Levula, H. Laakso, A. Ojala, O. Peltola, J. Heiskanen, J.
Pumpanen, and T. Vesala (2015), Carbon dioxide and energy fluxes over a small boreal lake in Southern
Finland, J. Geophys. Res. Biogeosci., 120(7), 1296-1314, doi: 10.1002/2014JG002873.
Saloranta, T.M., and T. Andersen (2007), MyLake – A multi-year lake simulation model code suitable for
uncertainty and sensitivity analysis simulations, Ecol. Model., 207(1), 45-60,
doi:10.1016/j.ecolmodel.2007.03.018.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Designing Very Shallow Water Bodies for Disinfection: Impact of
Daily Stratification/Destratification
C. J. Lemckert1*, N. W. Dahl2, P. L. Woodfield2, B.A.F. Simpson2, H. Zhang2, H. M.
Stratton3, A. Roiko4
1
Faculty of Arts and Design, University of Canberra, ACT, Australia
School of Engineering, Griffith University, Gold Coast, Queensland, Australia
3
School of Natural Sciences, Griffith University, Gold Coast, Queensland, Australia
4
School of Medicine, Griffith University, Gold Coast, Queensland, Australia
2
*Corresponding author, e-mail charles.lemckert@canberra.edu.au
KEYWORDS
Shallow water bodies; stratification; turbulent modelling; disinfection.
EXTENDED ABSTRACT
Introduction
Very shallow fresh water bodies are very common throughout the world. They occur as
both natural and constructed systems, with the latter being used extensively in waste water
treatment systems in the form of maturation ponds. These ponds, which are normally 0.9 to
1.5 m deep are used primarily for pathogen inactivation/removal by various mechanisms,
including naturally supplied sunlight that results in ultraviolet disinfection, predation and high
pH levels (Stratton et al., 2015). The treatment efficiency, with regard to pathogen removal is,
however, extremely variable between maturation ponds, which in part is due to a lack in the
understanding of the flow and mixing dynamics in these ponds.
While sunlight disinfection is an important component of the treatment, the vertical
movement of the pathogens is also critical. This movement is affected by the diurnal
stratification cycle, which is prevalent in these common, very shallow water systems.
Turbulent and thermal convection play a major role in transporting pathogens into the nearsurface region to be affected by the ultraviolet component of sunlight. This study analyses a
slice of a maturation pond and simulates E. coli moving within the slice and being affected by
the stratification, sunlight attenuation and mixing driven by wind shear and natural
convection. Importantly, the results found here are readily applicable to our understanding of
how pathogens can decay in natural shallow water systems.
Materials and methods
Measurements of various parameters, including temperature at various depths, solar
radiation and light attenuation within the water column, wind speed and direction and
pathogen levels, were taken in an operational maturation pond, fitted with baffles, located in
South-East Queensland (SEQ). Fig. 1a shows the pond layout with five baffles at
approximately 80% length. The average depth of the pond is 0.8 m with an average inflow of
10-3m3s-1 making the theoretical residence time ~16 days (Stratton et al., 2015).
The pond was modelled using the computational fluid dynamics (CFD) package ANSYS
FLUENT. A two-dimensional simulation in the vertical-horizontal plane of the first baffled
area was considered to be representative of the flow dynamics of the pond. Five different
turbulence models were considered for closure of the momentum conservation equations
Buoyancy effects were included in each turbulence model. Scalar transport, representative of
2
Physical Processes in Natural Waters 2015
E. coli, was simulated within the geometry. UV disinfection (Nguyen et al., 2015) of E. coli
was accounted for via the source term in the model equations.
Boundary conditions at the air-water surface included the shear stress from the wind,
shortwave and longwave radiation, and sensible and evaporative heat fluxes. Generation of
thermal energy from the attenuation of shortwave radiation in the water column was also
modelled. Unsteady simulations were done over a diurnal cycle with boundary conditions and
internal sources being applied from measured data.
a) Measured pond geometry
b) Pond outlet log-reduction of E. coli
Fig. 1. a) Measured geometry of the maturation pond - the water-level and baffle placement
are shown by dark solid lines. Note that the depth scale is exaggerated; b) Measured and
predicted outlet surface concentrations of E. coli for 6/7th March 2015.
Results and discussion
A key conclusion from this work is that the choice of turbulence model for CFD
simulations is less significant than the inclusion of the buoyancy production term in the
turbulent closure model. It has a major influence on thermal distributions, velocity
distributions and on the die-off of E. coli resulting from sunlight disinfection. As shown in
Fig. 1b, reasonable agreement is found between experimental and modelled die-off at midday.
Differences in the late afternoon can be explained by the neglect of other types of die-off
mechanisms (e.g. predation and starvation) and secondary effects (e.g. temperature, pH, ROS,
3D mixing). Improvements to modelling of shallow water systems, including maturation
ponds, could be made through further development and refinement of the turbulent buoyancy
production term (due to its pronounced influence on diurnal stratification/destratification) and
through greater elucidation of the various and complex mechanisms for die-off.
REFERENCES
Dahl, N.W., Woodfield, P.L., Lemckert, C.J., Stratton, H. and Roiko, A. (2017) A practical model for sunlight
disinfection of a subtropical maturation pond. Water Research 108, 151-159.
Nguyen, M.T., Jasper, J.T., Boehm, A.B. and Nelson, K.L. (2015). Sunlight inactivation of fecal indicator
bacteria in open-water unit process treatment wetlands: Modeling endogenous and exogenous inactivation
rates. Water Research 83, 282–292.
Stratton, H., Lemckert, C., Roiko, A., Zhang, H., Wilson, S., Gibb, K., van der Akker, B., Macdonald, J.,
Melvin, S., Sheludchenko, M., Li, M., Xie, J., Padovan, A. and Lehmann, R. (2015) Validation of
maturation ponds in order to enhance safe and economical water recycling. Australian Water Recycling
Centre of Excellence.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Comparison of ice drift characteristics derived from Eulerian and
Lagrangian measurements in the Gulf of Finland, the Baltic Sea
M-J. Lilover1, T. Kõuts1 and M. Leppäranta2
1
Department of Marine Systems at Tallinn University of Technology
Akadeemia tee 15A, 12618 Tallinn, ESTONIA
2
Department of Physics, University of Helsinki, Box 68 (Gustaf Hällströmin katu 2b),
Fi-00014 Helsinki, Finland
*Corresponding author, e-mail madis.lilover@ttu.ee
KEYWORDS
Gulf of Finland; ice drift measurements; ice dynamics; ice drift velocity spectra.
EXTENDED ABSTRACT
Introduction
The Gulf of Finland is of an elongated form – 330 km long and 80–100 km wide with the
main waterways extending along the Gulf for cargo vessels and across the Gulf for the
passenger ship traffic. The presence of ice cover sets specific navigation limits for as in many
cases cargo vessels need icebreaker assistance. Operating simultaneously several vessels in
the abovementioned circumstances requires the knowledge of local ice dynamics and
therefore we focused our study on this issue in the frame of EU project SAFEWIN. The study
aims to describe the high resolution sea ice dynamics at three different locations of the cross
transect of Gulf of Finland in relation to wind forcing using both ice drift Eulerian and
Lagrangian measurements supported by satellite remote sensing ice coverage maps. This kind
of measurements are unique in the Baltic Sea in general as well as in the Gulf of Finland in
particular.
Materials and methods
The winter 2010 was colder than normal and the whole Gulf of Finland froze over. In the
southern coast the ice conditions were quite variable due to south–westerly winds
occasionally producing large open water areas. The main study area was selected as the
section between Kotka, Finland and Kunda, Estonia, almost north–south in the longitude
section 26–27°E. MODIS (Moderate resolution imaging spectrometer of Aqua/Terra satellite
of NASA) images of the ice conditions at the study area were collected. Ice drift velocities
were measured by bottom-mounted ADCP and drifting buoys. The ADCP was deployed in the
central zone of the Gulf of Finland on 12 January 2010 and recovered on 27 April 2010. The
site (59°42.09’N, 26°24.23’E; depth 63 m) was located in the deep basin extending south–east
toward Kunda Bay, at around 15 km from the coast. A 307.2 kHz broadband ADCP
(Workhorse Sentinel, RD Instruments) deployed on the bottom was used. In the present ice
dynamics study, the bottom-track (BT) option of the ADCP with a sampling interval of 10
minutes (an average of five high-frequency pings) was utilized to trace from below the sea
surface/ice bottom. Thus the data include the ice drift velocity (its BT velocity), the BT error
velocity and the vertical profiles of current velocity with vertical resolution of 2 m (from 6 to
58 m). The drifters were launched on the 8th of March in two groups: #5 and #6 in the middle
of the gulf at about 15 nautical miles to the north of the ADCP station, with a maximum
distance between buoys of about two nautical miles. Drifters #8 and #10 were launched even
further to the north at around 32 nautical miles from the ADCP station (figure 1). The drifters
Physical Processes in Natural Waters 2015
2
are compact devices with the length of about 1 m, the diameter of 11 cm and the mass of 10
kg. Their measurement interval can be controlled remotely between 15 minutes and 2 hours
during field experiments. The ice drifter experiments lasted for total of 66 days until the 14th
of May. However, for this paper only the measurements performed until the end of March
were analysed as in April at the ADCP location the sea was free of ice.
Figure 1. MODIS image of the ice conditions in the Gulf of Finland on 9 March 2010, one day after to the
installation of the drifter buoys. Image corresponds to the ADCP ice-covered period C (origin of the image is
NASA, processed by Dr. Liis Sipelgas). The drifters are numbered as 5, 6, 8 and 10 and their positions are
marked with red dots, the ADCP location is marked with a yellow dot.
Results and discussion
The analysis of the experiment data revealed frequent alternation of ice cover periods
with open water periods in the Gulf of Finland. Leads were present on the north and south
coasts due to repeated changes in wind direction. A total of five ice periods with a duration of
6–12 days were obtained at the ADCP measurement site (site of Eulerian measurements) until
the end of March (periods A (Jan.), B (Feb.), C, D and E (all in Mar.). The ice drifting speed
at the ADCP site was well correlated with the wind data measured nearby. The sub-periods
mean asymptotic wind factor and deviation angle at large wind speeds were 0.034 and 9°,
respectively, in the oceanic boundary layer, corresponding to the ratio of 0.92 of the air–ice
and water–ice drag coefficients. The latter probably was affected by the geometry of the Gulf
of Finland. The asymptotic wind factor was higher than it is usually estimated but still in
accordance with other ice studies in the Baltic Sea (Björk et al., 2008; Leppäranta, 2005).
In period C three instruments (buoy 5, buoy 6 and the ADCP) showed a coinciding ice
drift direction, indicating homogenous ice drift in the middle of the Gulf of Finland. Still
closer to the south coast (ADCP site) for all 3 sub-periods C, D and E the mean ice drift speed
was substantially higher than in the central region measured by drifters (Lagrangian
measurements).
The clockwise (CW) spectra of fluctuating part of ice velocity (ice velocity minus mean
ice velocity) showed a wide peak at the inertial frequency and a power law with the exponent
of -1.9 for the higher frequencies. Eulerian CW spectra showed a higher level than the
Lagrangian spectra in the frequency range of 0.04–0.2 cph (the corresponding time periods 5
to 25 hours). Eulerian and Lagrangian counter-clockwise spectra had roughly the same energy
density values. Spatial correlations were significant in the range from 3 to 60 km, with the
best-fit falling power of -0.18. The correlation level approached 1 at about 1 km distance (i.e.,
at the scale of the size of ice floes, although the scatter of the correlation vs. length scale was
large). The estimated integral correlation length scale was 48 km.
REFERENCES
Björk, B., C. Nohr, B. G. Gustafsson, and A. E. B. Lindberg (2008), Ice dynamics in the Bothnian Bay inferred
from ADCP measurements, Tellus, 60A, 178-188.
Leppäranta, M. (2005), The Drift of Sea Ice, Springer, Helsinki.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
An integrated process-based minimal model to account for the
feedbacks between ecological and physical processes in lakes
G. López Moreira1,2*, M. Toffolon1, Franz Hölker2
1
Department of Civil, Environmental and Mechanical Engineering
University of Trento, Trento, Italy
2
Department of Ecohydrology
Leibniz-Institute of Freshwater Ecology and Inland Fisheries, Berlin, Germany
*Corresponding author, e-mails ga.lopez@unitn.it / ga.lopez@igb-berlin.de
KEYWORDS
Lake ecosystems; integrated model; process-based model; minimal model; feedbacks.
EXTENDED ABSTRACT
Introduction
Since their introduction in the 1960s, the development of ecosystem models has been
driven by the wide spectrum of potential applications they offer us to better reconstruct and
interpret experimental results, improve our understanding of the real world and allow us to
test hypotheses and predict future environmental changes under different scenarios such as
climate change. These models often fall into one of two types: overly simplistic
representations of isolated processes with limited potential to explain real-world observations;
or overly complex models that can hardly improve scientific understanding of the represented
system due to their results being too difficult to analyse in terms of fundamental processes
and controls. Over-parameterisation is also an issue in the latter case. Moreover, most models
are designed for very specific purposes and the usual way to cope with the limitations of a
single model is to partially couple it to another one, i.e., to feed the output of the one to the
other. This strategy hinders the consideration of the feedbacks that processes modelled by the
latter may give to processes modelled by the former, which is normally the case when a
hydrodynamic model is partially coupled to an ecosystem model.
However, ecological processes are now known to have the potential to significantly alter
the physical behaviour of aquatic ecosystems. For example, a higher light attenuation due to
increasing concentrations of dissolved organic carbon in natural waters, one of many
alterations associated to climate change, results in different physical responses of such
systems to solar radiation (Persson & Jones, 2008; Rinke et al., 2010). Through a similar
mechanism, planktonic events can alter the duration of the stratified period, thermal structure
and mixing regime of medium-depth temperate lakes (Shatwell et al., 2016). Yet feedbacks
are related not only to an altered underwater light climate, but to changes in other physical
properties as well. In this respect, studies have shown that exopolymeric substances secreted
by phytoplankton can modify the rheological properties of water to the extent of altering, for
instance, the thickness of a pycnocline (Jenkinson & Sun, 2011).
An integrated process-based minimal model to account for the feedbacks
With the aim of addressing the limitation of partially-coupled models to account for the
feedbacks between ecological and physical processes in lake ecosystems in general, we
further develop the model reported by Jäger et al. (2010). State variables are described by a
set of partial differential equations (PDEs) that are simultaneously solved while keeping
Physical Processes in Natural Waters 2017
2
model complexity to a minimum so that the relative importance of the different processes can
still be assessed, as well as that of the feedbacks between them.
Coefficients in these PDEs are dynamically calculated via another set of process-based
algebraic equations that depend on parameters whose values are calibrated within ranges
reported in the scientific literature, and on the state variables. Thus, all processes are fully
coupled and all known feedbacks are considered. External forcing can also be dynamically
simulated so that the response of the system can be assessed not only under steady state
conditions, but also under different scenarios involving the variation at different time scales of
solar radiation, air temperature, wind speed, precipitation, nutrient loading and
brownification, among others. The model also allows for the implementation of different
boundary conditions such as lake-groundwater exchanges.
Physical processes include light attenuation as a function of water colour and turbidity,
thermal stratification (and the associated density stratification) resulting from the absorption
of shortwave solar radiation, heat exchanges with the atmosphere and the lakebed, and the
turbulent diffusion of heat that is due to wind shear on the surface, accounting for the
reduction of the mixing under stratified conditions by an empirical relationship based on the
Richardson number.
Ecological processes include light- and nutrient-limited primary production and nutrient
cycling. Modelled primary producers can potentially comprise multiple taxonomic groups of
phytoplankton, periphyton and macrophytes, making it possible to analyse how community
structure may change, for example, under climate change scenarios (as in Vasconcelos et al.,
2016), where the temperature dependence of growth, metabolic and other biological rates
plays an important role. Nutrients are cycled between the water column, the sediments, where
remineralisation of organic nutrients occurs, and the biotic components of the system. For the
latter, nutrient uptake rates are modelled as a function of the concentration of dissolved
nutrients, the corresponding nutrient quotas, which can be either fixed or dynamically
simulated, and potentially water temperature. Finally, predation pressure exerted on primary
producers can also be incorporated as a fixed or variable component of the loss rates.
Expected impact
Relatively few modelling studies consider these feedbacks, and those that do usually
focus on the very specific lake types for which enough data is available. Therefore, their
actual importance for a more accurate representation of lake ecosystems in general is still
unknown. By applying the proposed model to a wider spectrum of lake types, particularly to
those for which high-frequency data is now being produced (e.g. shallow subtropical lakes),
we expect to gain new insights on the fundamental processes driving the overall functioning
of these ecosystems and the potential effects of changing environmental factors.
REFERENCES
Jäger, C. G., S. Diehl, and M. Emans (2010), Physical Determinants of Phytoplankton Production, Algal
Stoichiometry, and Vertical Nutrient Fluxes, The American Naturalist, 175(4), E91-E104,
doi:10.1086/650728
Jenkinson I. R., and J. Sun (2011), A model of pycnocline thickness modified by the rheological properties of
phytoplankton exopolymeric substances, Journal of Plankton Research, 33(3), 373-383,
doi:10.1093/plankt/fbq099
Persson, I., and I. D. Jones (2008), The effect of water colour on lake hydrodynamics: a modelling study,
Freshwater Biology, 53, 2345-2355, doi:10.1111/j.1365-2427.2008.02049.x
Rinke, K., P. Yeates, and K. Rothhaupt, (2010), A simulation study of the feedback of phytoplankton on thermal
structure via light extinction, Freshwater Biology, 55, 1674-1693, doi:10.1111/j.1365-2427.2010.02401.x
Shatwell, T., R. Adrian, and G. Kirillin (2016), Planktonic events may cause polymictic-dimictic regime shifts in
temperate lakes, Scientific Reports, 6, 24361, doi:10.1038/srep24361
Vasconcelos, F. R., S. Diehl, P. Rodríguez, et al. (2016), Asymmetrical competition between aquatic primary
producers in a warmer and browner world, Ecology, 97(10), 2580-2592, doi:10.1002/ecy.1487
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Hydrodynamic characteristics of the Xiangxi Bay in the Three
Gorges Reservoir, China
Jun MA1, Zheng-Jian YANG1, Dao-Bin JI2, De-Fu LIU1*
1
Hubei Key Laboratory of Ecological Restoration of River-lakes and Algal Utilization,
Hubei University of Technology, Wuhan, China
2
Engineering Research Center of Eco-environment in Three Gorges Reservoir Region,
Ministry of Education,
China Three Gorges University, Yichang, China
*Corresponding author, e-mail dfliu@189.cn
ABSTRACT:
Based on continuous monitoring in 2010, this paper analysed the hydrodynamic processes of
the Xiangxi Bay(XXB), a tributary of Three Gorges Reservoir (TGR). It showed that the
hydrodynamics of the XXB could be generalized as a density-stratified flow, and could not be
simply simulated by one-dimension model. The upstream water mainly flowed out of the bay
in a process of downslope-bottom density current, meanwhile, the TGR mainstream water
entered into the XXB as a reverse density current. The density current of the XXB was mainly
caused by the water temperature difference and turbidity difference between the TGR
mainstream and the bay, and the former played a dominant role.
KEYWORDS
Three Gorges Reservoir; Xiangxi Bay; hydrodynamics; density current; water temperature difference.
EXTENDED ABSTRACT
Introduction
The reservoir behind the controversial Three Gorges Dam (TGD) — the world’s largest
hydropower station(Stone, 2008), located on the Yangtze River in Hubei Province, China —
has been regulated with the top water level at 175 m above sea level since October 26, 2010.
The changed hydrological conditions caused serious impacts on the environment and
ecosystem (Fu et al., 2010; Shen and Xie, 2004; Stone, 2008). One of the most severe
challenges, algal blooms impairing the aquatic ecosystem, and threatened drinking water
quality and human health (Liu et al., 2012), has become a serious social and environmental
problem for the TGR(Wu, 2008.).
As no blooms happened before the construction of TGD, the inducement for blooms
could certainly be attributed to the TGD (Fu et al., 2010). Undoubtedly, the dam can only
directly change the hydrology, then indirectly influence the eco-environmental parameters
condition and cause some problem such as algal blooms. In practice, the hydrodynamic in the
TGR present not a simple one-dimensional flow from upstream to downstream, especially, in
some tributary bays, complex bidirectional currents dominate(Holbach et al., 2014; Ji et al.,
2010), which are different from some other reservoirs in the world. Thermal stratification(Yi
et al., 2009)and vertical mixing(Liu et al., 2012) caused by those currents might having great
effect on the presence or absence of algal blooms. In this present study,we used a special
method to investigate of hydrodynamic, in order to study the mechanism of the algal bloom in
the tributary bays of TGR.
Physical Processes in Natural Waters 2017
2
Materials and methods
Sampling sites
The Xiangxi River (XXR) is the largest tributary close to the TGD in Hubei Province
(Fig. 1 b), which is approximately 94 kilometers long and located in a subtropical continental
monsoon climate. The average annual temperature is 16.6 °C, and annual rainfall and river
discharge are 1015.6 mm and 40.18 m3/s, respectively. When the TGR operated at a water
level of 175 m, a 40 km reach covered by the backwater from the estuary, called as Xiangxi
Bay (XXB). Since the initiation of storage in 2003, different phytoplankton species dominate
XXB during different seasons(Liu et al., 2012; Wang et al., 2011).
In order to monitor the temporal and spatial variation of the eco-environmental
parameters, we set eleven sampling sites (at intervals of approximately 3 km) in the XXB,
indicated as XX00-XX10 (Fig. 1 c) in succession from the estuary to the end of the
backwater. Another site was located at the mainstream of the Yangtze River to represent the
mainstream of the reservoir, indicated as "GJB" (Fig. 2). A site was also located in the inflow
river of the XXB, indicated as "Inflow" (Fig. 2).
Fig. 1 (a) Location of the Three Gorges Reservoir (TGR) in China; (b) Location of Xiangxi
Bay (XXB) in the TGR outlined in red; and (c) Location of the sampling sites in the XXB,
where XX00 is near the confluence with the TGR
Measurements of velocity.
Profiles of velocity was measured from a boat equipped with an Acoustic Doppler Vector
velocimeter (ADV; Nortek, Norway). Vertical resolution of the measurements was 1 m.
Using the method proposed by Ma (Ma et al., 2011), flow velocities in east (Ve), north (Vn)
and vertical (Vu) directions were measured at depths of 0.5 and 1 m and then to the bottom at
1 m intervals.
Hydrodynamic characteristics of the Xiangxi Bay in the Three Gorges Reservoir, China
3
Fig. 2 Schematic diagram of field monitoring of velocity(Ma et al., 2011)
Results and discussion
Hydrodynamic process and characteristics of the Xiangxi Bay (XXB)
Overall, the flow rate of the Xiangxi Bay (XXB) is small and the average flow rate is
only centimeter level. The flow direction of the XXB is from north to south, and the direction
of the water flow is roughly the same as that of the XXB, and the lateral velocity is small.
Therefore, when analyzing flow characteristics of the XXB, the northward divergence
velocity vector was used.
Fig. 3 is the dynamic distribution of the longitudinal velocity of the XXB in 2008. It can
be seen that the water flow is not a 1-D flow but a stratified flow during the whole year.
In the reservoir water supply, water level drawdown and flood season, the upper reaches
water flows to the estuary from the bottom of the XXB. The bottom flow velocity is
increasing since May. The maximum upstream velocity reaches 0.46 m/s (Fig. 3 (h)) on June
22nd, which is related to the sudden increase in inflow at the upstream in the flood season.
In the upper reaches of the XXB, water flows to the estuary from the bottom since the
middle of February. Meanwhile, there is a "wedge" shaped water flowing into the XXB at the
bottom of the XXB estuary from the Yangtze River. The reverse density current travels about
5 km from the estuary and the average flow rate is 0.04 m /s (Fig. 3 (b)). The travel distance
gradually expanded to about 7 km from the estuary in March (Fig. 3 (c)).
There is a continuous stratification flow in the XXB since then. Water intrudes from the
Yangtze River into the XXB at the middle layer of the estuary range from 5-40 m depth from
April to September. The average flow rate is 0.05 m/s and the maximum velocity reaches to
0.16 m/s at 5 m depth on May 10 (Fig. 3 (f)). The travel distance of stratification flow from
April to September is significantly further than that of February and March. The reverse
density flow can affect the upper reaches of the XXB most of the time from April to
September.
During the first stage of reservoir impoundment from the end of September to October
6th, the density current intrudes into the XXB from the surface and immediately affects the
entire XXB. The average flow rate was 0.06 m/s (Fig. 3 (k), (l)). But the density current
intruded from the middle layer during the second impoundment (Figure 3 (m), (n)). At the
same time, the upper reaches water of the XXB still flows out to the estuary. There is still
4
Physical Processes in Natural Waters 2017
-10
-20
-30
-40
-50
-60
-70
-10
-20
-30
-40
-50
-60
-70
-5
-15
-25
-35
-45
-55
-65
-75
(g)
0
0.391
May.24,2008
0
0.462
Jun.22,2008
(h)
(j)
(i)
0
0.278
Jul.10,2008
0
0.454
Aug.17,2008
(l)
(k)
0
0.229
0
0.183
Sept.30,2008
0
0.184
(m)
Oct.19,2008
0
0.251
Oct.03,2008
(n)
Oct.25,2008
-5
-15
-25
-35
-45
-55
-65
-5
-15
-25
-35
-45
-55
-65
-75
-5
-15
-25
-35
-45
-55
-65
-75
-15
-15
-30
-30
-45
-45
-60
(p)
(o)
-60
-75
0
Dec.11,2008
Nov.25,2008
0.464
-90
-90
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28
距 河 口 的 距 离 (km)
距 河 口 的 距 离 (km)
-75
0
0.245
Distance from the estuary (km)
(m)
水 深(m)
Depth
(m)
水深
Depth
(m)
水 深 (m)
Depth
(m)
May.10,2008
水 深 (m)
Depth
(m)
Apr.19,2008
0
0.217
(m)
水 深(m)
Depth
-5
-15
-25
-35
-45
-55
-65
(f)
(e)
0
0.345
-10
-20
-30
-40
-50
-60
-70
-5
-15
-25
-35
-45
-55
-65
水深
Depth
(m)(m)
-10
-20
-30
-40
-50
-60
-70
(m)
水 深(m)
Depth
距 河 from
口 的the
距 estuary
离 (km)(km)
Distance
the离estuary
Distance
距 河 口from
的 距
(km) (km)
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28
-5
-5
-15
-15
-25
-25
-35
-35
-45
-45
(b)
(a)
-55
-55
0
0
-65
Feb.16,2008 -65
Jan.17,2008
0.196
0.32
-75
-75
-5
-10
-15
-20
-25
-30
-35
-40
-45
-50
(c)
(d)
-55
-60
0
0
-65
Apr.12,2008 -70
Mar.15,2008
0.284
0.324
-75
(m)
水 深(m)
Depth
Depth
(m)
水 深(m)
水 深(m)
(m)
Depth
Depth
(m)
(m)
水深
Depth
水 深(m)
(m)
(m)
Depth
水 深(m)
Depth
(m)
水 深 (m)
Depth
(m)
水 深(m)
Depth
(m)
水 深(m)
density current intruding from the bottom or middle layer after impoundment in November
and December. But the intruding distance is less than 10 km (Figure 3 (o), (p)).
Thus, the hydrodynamics of the XXB are difficult to be generalized as one-dimensional
characteristics, but in the long-term, there are complex stratified density flows in the vertical
direction. In the whole year, there is always the underflow which flows to the estuary at the
bottom of the XXB, and density current intrusion will occur at the estuary most of the time.
The plunging point and travel distance of intruding density current will vary with the physical
properties of the water, inflow of the XXB, water level and water level daily fluctuation in
different seasons.
Distance from the estuary (km)
Fig. 3 Longitudinal-vertical velocity profile of the Xiangxi Bay in 2008 (Unit: m/s)
Hydrodynamic characteristics of the Xiangxi Bay in the Three Gorges Reservoir, China
5
Note: The red vector indicates the water flows into the XXB, the blue vector indicates the
water flows out of the XXB. The vector length indicates the magnitude of flow velocity.
REFERENCES
Fu, B., Wu, B., Lv, Y., Xu, Z., Cao, J., Niu, D., Yang, G., Zhou, Y., 2010. Three Gorges Project: efforts and
challenges for the environment. Progress in Physical Geography 34, 741-754.
Holbach, A., Norra, S., Wang, L., Yijun, Y., 2014. Three Gorges Reservoir: Density Pump Ampli fi cation of
Pollutant Transport into Tributaries. Environmental Science & Technology 48, 7798-7806.
Ji, D., Liu, D., Yang, Z., Xiao, S., 2010. Hydrodynamic characteristics of Xiangxi Bay in Three Gorges
Reservoir. Science China Technological Sciences 40, 101-112 (in Chinese).
Liu, L., Liu, D., Johnson, D.M., Yi, Z., Huang, Y., 2012. Effects of vertical mixing on phytoplankton blooms in
Xiangxi Bay of Three Gorges Reservoir: Implications for management. Water Research 46, 2121-2130.
Ma, J., Liu, D., Ji, D., Yang, Z., Yi, Z., 2011. Flow measurement methods under low flow velocity at the
tributary bays of Three Gorges Reservoir. Journal of Yangtze River Scientific Research Institute 28, 3034+54(in Chinese).
Shen, G., Xie, Z., 2004. Three Gorges Project: Chance and Challenge. Science 304, 681.
Stone, R., 2008. Three Gorges Dam: Into the Unknown, Scinece News. Scinece, pp. 628-632.
Wang, L., Cai, Q., Xu, Y., Kong, L., Tan, L., Zhang, M., 2011. Weekly dynamics of phytoplankton functional
groups under high water level fluctuations in a subtropical reservoir-bay. Aquatic Ecology 45, 197-212.
Wu, X., 2008. To Boost the Environment Protection by Perfecting the Standards of Environment Protection
(http://www.gov.cn/zxft/ft108/wz.htm). Ministry of Environmental Protection of the People's Republic of
China, Beijing (in Chinese).
Yi, Z., Liu, D., Yang, Z., Ma, J., Ji, D., 2009. Water temperature structure and impact of which on the bloom in
spring in Xiangxi Bay at Three Gorges Reservoir. Journal of Hydroecology 2, 6-11(in Chinese).
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
The Role of Water Level Fluctuation on GHG Dynamics in a
Temperate UK Reservoir
Roseanne McDonald1, 2*, K. Dinsmore1, M. Billett2, U. Skiba1, C. Evans1 and S. Waldron3
1
Centre for Ecology & Hydrology, United Kingdom
Biological and Environmental Sciences, University of Stirling, United
Kingdom
3
School of Geographical and Earth Sciences, University of Glasgow, United
Kingdom
2
*Corresponding author, e-mail rosdon23@ceh.ac.uk
KEYWORDS
Reservoir drawdown; sediment fluxes; aquatic carbon; biogeochemistry; inland waters.
EXTENDED ABSTRACT
Introduction
Inland waters play an important role in the transport, transformation, loss and storage
of carbon (C) in the pathway between terrestrial and marine systems. Reservoirs in particular
are likely to represent biogeochemical cycling hotspots, and a potential source of greenhouse
gases (GHGs) to the atmosphere due to large inputs of terrestrial carbon alongside operational
and mixing processes. Although all aquatic systems experience natural fluctuations in water
level, this is often more extreme in reservoir environments due to seasonal demands or
operational maintenance. Drawdown emissions occur when fluctuating water levels cause
changes in hydrostatic pressure and create sediments that are periodically inundated with
water and then exposed to aerobic conditions. During drawdown events, ebullition becomes
more intense and bubbles bypass methane consumption in the sediment and water column
leading to a larger atmospheric methane flux. Vegetation which re-colonises this zone may
also provide a significant labile C input on rewetting. As these processes can act to convert
atmospheric CO2 to the more potent GHG CH4, this zone may contribute disproportionately to
emissions from the system as a whole. Recent drawdown studies have found exposed
reservoir sediments to produce pulses of CH4 (Harrison et al., 2017) and CO2 (Jin et al., 2016)
emissions. Whilst many studies focus on the magnitude of reservoir emissions, we still lack a
clear picture of the internal controls on GHG production within the complex reservoir system.
Materials and methods
Bi-weekly samples were taken from Waltersmuir Reservoir over a one year period,
January 2016 to January 2017. During the 11-week drawdown event from April to July,
sampling frequency increased to weekly. Waltersmuir Reservoir is located in central Scotland
and has a surface area of 3 ha, mean depth of 4.4 m and catchment area of 1169 ha. Duplicate
water samples were collected at five sampling locations using a 60 ml syringe and filtered in
the field. Dissolved oxygen (DO), pH, temperature and electrical conductivity (EC) were also
measured in-situ. Water samples were analysed for DOC and DIC on a LabTOC instrument
(PPM Solutions), and ammonia, nitrate and nitrite using an AQ2 discrete analyser (Seal
Physical Processes in Natural Waters 2015
2
Analytical). Dissolved CO2, CH4 and N2O gas samples were collected using the headspace
technique which is fully described by Kling et al., (1991) and Billett et al., (2013). A 40 ml
water sample, collected from ~10 cm water depth, was equilibrated with 20 ml headspace of
ambient air in a 60 ml syringe by shaking vigorously underwater for one minute. The
headspace was then injected into a 12 ml Exetainer ® and analysed on a 7890B gas
chromatograph (Agilent Technologies) for CH 4, CO2 and N2O. During the 11-week
drawdown and rewetting period, fluxes at the sediment-water interface were measured weekly
in a transect using nine static and soil respiration chambers covering areas of different slope
and sediment moisture. To investigate the drivers of measured fluxes, water-table, water
chemistry and sediment nutrient concentrations were also analysed. A UAV survey also took
place during maximum drawdown to help quantify sediment evasion and upscale emissions.
Results and discussion
Dissolved GHG concentrations (Fig. 1), particularly CH4 (median = 0.62 μg C L-1; max =
393.3 μg C L-1) and N2O (median = 0.38 μg C L-1; max = 17.6 μg C L-1) show increased
concentrations during the drawdown period. Two CH 4 peaks coincide with two partial
refilling events during periods of heavy rainfall. Similar but delayed peaks are also seen in
N2O concentrations, likely linked to a reduction in anaerobic activity as water levels fall.
DOC and DIC both increase during the drawdown period but with greater peaks in late
Summer.
Figure 1 (left). Dissolved GHG concentrations from Waltersmuir reservoir from January
2016-2017 with grey area highlighting 11 week drawdown. Figure 2 (right). Sediment fluxes
across all nine chambers, with higher fluxes in wetter areas.
Sediment fluxes (Fig. 2) were low across chambers 1-7, with increased fluxes observed at
chambers 8 and 9 which were consistently saturated with water. A PCA was performed to
examine the covariances and correlations between variables, where 56% of total variation was
explained with the top three principal components. At the 95% confidence level, a statistically
significant relationship was found between CO 2 and soil moisture (p=0.0001); CO2 and pH
(p=0.0244); CH4 and pH (p=0.00601), and CH4 and soil moisture (p=0.006). The results
highlight that pulses of aquatic CH4 concentrations occur during rewetting, with sediment
fluxes linked to moisture and pH. Further studies are required to further understand spatial
and temporal variability of drawdown events on total reservoir C budgets.
REFERENCES
Billett, M. F. and F. H. Harvey (2013). Measurements of CO2 and CH4 evasion from UK
peatland headwater streams. Biogeochemistry 114, 165-181, doi: 10.1007/s10533-012-97989.
Arthur et al.
3
Harrison, J. A., B. R. Deemer, M. K. Birchfield, and M. T. O’Malley (2017). Reservoir
Water-Level Drawdowns Accelerate and Amplify Methane Emission. Environ. Sci. Technol.,
51 (3), 1267-1277. doi: 10.1021/acs.est.6b03185.
Jin, H., T. K. Yoon, S. Lee, H. Kang, J. Im and J. Park (2016). Enhanced greenhouse gas
emission from exposed sediments along a hydroelectric reservoir during an extreme drought
event. Environ. Res. Lett, 11, 124003. doi: 10.1088/1748-9326/11/12/124003.
Kling, G. W., G. W. Kipphut, and M. C. Miller (1991). Arctic streams and lakes as conduits
to the atmosphere: Implications for tundra carbon budgets. Science 251, 298-301. doi:
10.1126/science.251.4991.298.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Light and hydrodynamics as key drivers behind the recent decline
of Planktothrix rubescens in a mesotrophic lake (Lake Hallwil)
C. Ahnlund-McElgunn1, N. Gallina1, A. Stöckli2, B. Ibelings1, D. F. McGinnis1,*
1
Aquatic Physic Group, Department F.-A. Forel (DEFSE), Faculty of Science, University of
Geneva, Boulevard Carl Vogt 66, 1211, Geneva, Switzerland
2
Department of Civil Engineering, Transportation and Environment, Aarau, Switzerland
*Corresponding author, e-mail daniel.mcginnis@unige.ch
KEYWORDS
Planktothrix rubescens; environment; water quality; modelling; internal waves.
EXTENDED ABSTRACT
Introduction
After the onset of eutrophication in the mid-1900’s, substantial efforts were implemented
at Lake Hallwil (Canton Aargau, Switzerland) to reduce nutrient loading (Fig. 1). The
aggressive restoration program was extremely successful, and resulted in returning Lake
Hallwil from nearly hypereutrophic to now meso-/oligotrophic conditions. With the phosphorus
reduction and improvement, the nuisance cyanobacteria Planktothrix rubescens began to
reappear around 1985. The species had previously lived in Lake Hallwil at turn of the 20th
century, but had disappeared with the onset of eutrophication in the mid-1900s.
The cyanobacteria Planktothrix rubescens have been the dominant algal species in Lake
Hallwil since about 1999 and peaked in concentration in 2002. This red filamentous nuisance
species resides in the metalimnion and utilizes gas vesicles to achieve neutral density at optimal
light and nutrient levels. These restoration efforts have led to a decrease in most phytoplankton
species (Fig. 1) since 2000, resulting in increased Secchi disk depth and allowing light to access
the metalimnion where the P. rubescens thrive. Consequently, an increase in biomass of the
species was observed as the trophic state improved from 1985 - 2002. Since 2002, however, the
P. rubescens populations have continued to decline. In this paper, we explore the relationship
between their sustainability in relationship to local hydrodynamics.
Algae Wet Weight (g m-2)
80
70
SUM
CYANO.
CYANO(%)
TP
60
50
40
150
100
30
50
20
10
TP Yearly Average (mg m-3)
200
90
0
0
1985
1990
1995
2000
2005
2010
2015
Fig. 1. P. rubescens (CYANO) compared to total algae (SUM). Black open symbols:
% of P. rubescens compared to total biomass. Blue open symbols: yearly average
total P (The Department of Bau, Verkehr und Umwelt, Abteilung für Umwelt, Canton
Aargau).
2
Physical Processes in Natural Waters 2017
Methods
Lake Hallwil is glacially formed, with a mean depth of 28 m. Located in the central Swiss
plateau, the lake has undergone artificial aeration since 1985 (McGinnis et al., 2004).
Biogeochemical data were provided by the Department of Bau, Verkehr und Umwelt, Abteilung
für Umwelt, of the Canton Aargau. A temperature mooring was installed to resolve the basinscale diffusivities (Kz) with the heat budget method (Wüest et al., 2000). The water column
stability is calculated as N2 = gp-1∂ρ/∂z (s-2), while the Osmidov Length Scale is 𝐿2𝑜 =
𝐾𝑧 /(0.15𝑁) (m2).
Results and discussion
Secchi depths increased from ~2 m in 1999 to 6 m in 2015/2016. Though decreasing since
their peak in 2002, in 2015 P. rubescens began a more dramatic decline in concentration. Our
results show that P. rubescens follow the depth where light penetration is ~0.1 - 1% of surface
irradiance, and now reside at the base of the metalimnion (Fig. 2). As Lake Hallwil experiences
regular basin-scale motions (seiche) (McGinnis et al., 2004), the lower boundary of the
metalimnion is characterized by increased turbulent mixing length scales and decreasing water
column stability (Fig. 2, right). Therefore, when P. rubescens become deeper (Fig. 2 center),
they experience increasingly larger eddy length scales which exceed their daily migration rate
and thus transport P. rubescens into the hypolimnion at an unsustainable rate (Fig. 2).
As their migration (buoyancy compensation) rate is 1 m d -1, the combination of light
penetration and hydrodynamics in Lake Hallwil therefore suggest that P. rubescens can only
inhabit the metalimnion (~ 5 – 14 meters), with their position determined by the light
availability. Above or below this layer, the turbulent length scales become too large and stability
too low, overcoming their buoyancy compensation ability. The results of this study therefore
provide important insights into management strategies to mitigate P. rubescens in lakes and
suggest in the case of Lake Hallwil, a further improvement in Secchi depth will force P.
rubescens into an unsustainable habitat.
Stability N2 (s-1)
1E-5 1E-4 1E-3
Depth (m)
0
10
12.3 m
13.4 m
20
22 June, 2016
20 July, 2016
30
5
10
15
20
Temperature (°C)
0
5
10 15 20 0.1
Turbidity (NTU)
1
10
Max. Eddy (m)
Fig. 2. Left: Temperature profile from summer 2016.
Center: P. rubescens distribution in June and July. Right:
Mean N2 and Lo (estimated from basin-scale diffusivities).
REFERENCES
McGinnis D.F., Lorke A., Wuest A., Stockli A., Little J.C. (2004) Interaction between a bubble plume and the
near field in a stratified lake. Water Resources Research, 40, 11.
Wüest A., Piepke G., Van Senden D.C. (2000) Turbulent kinetic energy balance as a tool for estimating vertical
diffusivity in wind-forced stratified waters. Limnol. Oceanogr., 45, 1388-1400.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Effect of density currents on the seasonal evolution of basin-scale
internal waves in a Tropical Andean reservoir
A. Posada-Bedoya1*, A. Gómez-Giraldo1, R. Román-Botero1
1
Department of Geosciences and Environment, Universidad Nacional de Colombia, Medellin,
Colombia
*Corresponding author, e-mail afposadab@unal.edu.co
KEYWORDS
Tropical Andean reservoir, internal waves, density currents, seasonal evolution.
EXTENDED ABSTRACT
Introduction
Existing literature about basin-scale internal waves in temperate lakes and reservoirs is
prolific. However, there are few investigations about tropical Andean lakes, where thermal
stratification is typically weak, unlike the strong thermal gradients observed during summer in
temperate systems. Similarly, the interaction between different processes commonly observed
in stratified flows has been little addressed, such as is the case of the interaction between
density currents and basin-scale internal waves. Recent investigations have shown that density
currents are a prominent feature for the seasonal evolution of the thermal structure in tropical
Andean reservoirs (Román-Botero et al., 2013), which in turn should rule basin-scale internal
waves behaviour. Motivated by the lack in this topic, we investigated, at a seasonal scale, the
effect of the inflow hydrological regime on the internal wave field in a tropical Andean
reservoir.
Materials and methods
The study site was Porce II reservoir, located at 06°47’00”N in Colombia. Porce II is an
elongated canyon reservoir, 9.8 km long, with a mean width of 1 km and maximum depth of
92 m. The reservoir was continuously monitored for 170 days during wet and dry seasons
with two thermistor chains and a meteorological station. Inflow and outflow discharges were
provided by the reservoir managing company, Empresas Públicas de Medellín, and inflow
temperatures were recorded during most part of the survey. Wavelet analysis was conducted
on thermistor data to determine periodicity and vertical structure of the internal wave field.
Theoretical natural modes were calculated every ten days through the survey by solving the
2D eigenmodel proposed by Fricker and Nepf (2000). Intrusion depth of inflows was
estimated by solving the inflow mixing model described in Fischer et al. (1979).
Results and discussion
Significant differences in the basin-scale internal wave field were observed during the
different hydrological stages (Fig. 1). Transitional and dry stages were dominated by V1H1
and V2H2 natural modes, respectively (Fig. 1d,e); both excited by resonance with the diurnal
wind forcing (Fig. 1c). By contrast, despite the fact that during the wet season theoretical
V1H1 natural period matches the diurnal frequency of the wind (Fig. 1d), the internal wave
signature in the temperature records disappears below 20 to 30 meters depth, as is shown by
the continuous wavelet analysis of coherence and phase around the 24-h period band
conducted between temperature signals recorded at the different depths (Fig. 1e). This
Physical Processes in Natural Waters 2017
2
behaviour is a consequence of the inflows density currents, whose intrusion depth is highly
variable during this season (Fig. 1f) hence destroying coherent motions in the water column.
The high variability in the intrusion depth is associated to the small temperature gradient of
the water column during the wet season (Fig. 1f) with a top to bottom temperature difference
of 2°C (Fig. 1b), close to the amplitude of the diurnal cycle of temperature of the inflow (Fig.
1b). During the transition and dry seasons, natural modes can be excited by the wind given
that inflow discharges are low and confined above the first 20 m (Fig. 1f).
Figure 1. Seasonal evolution of external forcings and internal waves. (a) Flow, (b) inflow,
surface and bottom temperature, (c) wavelet power spectrum of wind speed, (d) vertically
integrated wavelet transform of discretized potential energy (PE) signal (contours) and natural
periods evolution (black lines), (e) 24-h period wavelet coherence (contours) and phase
difference (arrows) between temperature recorded at every thermistor against temperature at
12-m depth; right-pointing arrows (→) indicate in-phase fluctuations and left-pointing arrows
(←) anti-phase fluctuations. (f) Thermal structure (contours), intrusion depth of inflows
(white continuous line) and outtake depth (white dotted line).
REFERENCES
Fischer, H. B., List, E. J., Koh, R. C. Y., Imberger, J., and Brooks, N. H. (1979). Mixing in inland and coastal
waters. Academic.
Fricker, P. D. and Nepf, H. M. (2000). Bathymetry, stratification, and internal seiche structure. Journal of
Geophysical Research, 105(C6), 14237–14251.
Román-Botero, R., Gómez-Giraldo, E. A., and Toro, F. M. (2013). Seasonal effect of tributaries on the thermal
structure of a small Neotropical reservoir, La Fe – Colombia. Dyna. 80(177), 152-161.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
FVCOM modelling study of physical processes in a Scottish
fjordic system
B. Rabe1*
1
Marine Laboratory,
Marine Scotland Science, Aberdeen, UK
*Corresponding author, e-mail b.rabe@marlab.ac.uk
KEYWORDS
Modelling; fjords; estuarine circulation; FVCOM; Scottish sea loch.
EXTENDED ABSTRACT
Introduction
Scottish coastal waters are characterized by an intricate coastline and complex
bathymetry. The Scottish west coast consists of over 100 fjordic systems (sea lochs) with
different characteristics often separated into two or more basins by sills (Edwards and
Sharples 1986). A newly developed hydrodynamic model of the Scottish shelf, an FVCOM
implementation (Chen et al. 2003, Wolf et al. 2016), is used to evaluate and interpret
circulation patterns within one of Scotland’s largest sea lochs, Loch Linnhe. The circulation in
this system is forced by buoyancy gradients, winds, tides, and the Earth’s rotation. Loch
Linnhe has been extensively studied in the past with both models and observations because it
is an economically valuable system including a variety of ecosystem services.
Materials and methods
In addition to the wider Scottish shelf model, four smaller-scale case studies have been
implemented. These include higher resolution grids, especially in regions of interest, and the
case studies have been run for certain time periods in addition to the climatology run. One of
the case studies focuses on the wider Loch Linnhe system. We investigate a climatological
run as well as a run from May to October for the year 2011 as an example. This time period
was chosen because extensive field data from Loch Linnhe exists for that year along with
previous model runs by a different model (i.e. POLCOMS, Holt and James 2001).
Two comparisons are performed between the unstructured grid FVCOM model and the
previously run structured grid POLCOMS model. One focus is on differences in freshwater
input (leading to differences in vertical salinity structure) and grid differences, when each
model was forced with the best available forcing at the time the model was run. The other
focus is on using the same (or as similar as possible) forcing and boundary data and then
investigate differences between the two model outputs. A comparison against available
observational data will also reveal goodness of fit of the two different models.
Results and discussion
Comparisons of model differences independent of forcing and boundary data leads to the
following preliminary results:
x Freshwater input: Freshwater has a strong influence on sea lochs and coastal areas and
is represented in the FVCOM model through a daily freshwater input derived from a
distributed rainfall-runoff and routing model (Robson et al. 2010). In the POLCOMS
Physical Processes in Natural Waters 2017
2
model freshwater was introduced as distributed sources. These represented
contributions from only a few side lochs and might therefore be cruder.
x Model grid: The FVCOM model resolves islands, sills, and the open boundary much
better due to its flexible grid structure and smaller mesh size (tens of meters) in areas
of interest/importance (Fig. 1). The POLCOMS model has a fixed 100x100m
horizontal grid, and side lochs were not resolved. The vertical grids are different as
well (sigma layers versus an s-grid configuration).
An investigation of the circulation within this sea loch with the help of the two different
models (when using similar forcing and boundary data) looks at, for example, processes
caused by wind forcing. Physical processes, such as deep water renewal over the 12m deep
sill between the upper and the lower basins, is represented in the FVCOM model but not the
POLCOMS model. An estuarine circulation develops in the system, which causes for example
halocline differences in time and space, which in turn are of importance for ecological
processes.
Due to an increasing interest in the extension of existing ecosystem services of the Loch
Linnhe area and also new forms of development a good understanding of the underlying
physical processes of the system and its connectivity to other areas are required.
REFERENCES
Chen, C., Liu, H., & Beardsley, R. C. (2003). An unstructured grid, finite-volume, three-dimensional, primitive
equations ocean model: application to coastal ocean and estuaries. Journal of atmospheric and oceanic
technology, 20(1), 159-186.
Edwards, A. and F. Sharples (1986), Scottish sea loch catalogue. National Conservatory Council.
Holt, J.T. and I.D. James (2001), An s coordinate density evolving model of the northwest European continental
shelf, Model description and density structure, Journal of Geophysical Research – Oceans, 106, 1401514034.
Robson, A. J., S. J. Cole, and R. J. Moore (2010), Assessment of the G2G Model performance for the Cumbria
floods of 21 November 2009. Report to the Operational Implementation of Grid to Grid onto the NFFS
ProjectRep.
Wolf, J., N. Yates, A. Brereton, H. Buckland, M. De Dominicis, A. Gallego, and R. O’Hara Murray (2016), The
Scottish Shelf Model. Part 1: Shelf-Wide Domain, Scottish Marine and Freshwater Science, 7(3), 151pp.
Figure 1: Left: POLCOMS Loch Linnhe model bathymetry (no side lochs resolved). Small map shows location
of Loch Linnhe. Right: FVCOM model grid showing the larger extent of this case study and Loch Linnhe
(including side lochs) in the black box.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Vertical mixing in a tropical Andean Reservoir, Porce II
R. Román-Botero1*, L. Boegman2, A. Gómez-Giraldo1
1
Department of Geosciences. Universidad Nacional de Colombia, Medellín, Colombia
2
Department of Civil Engineering. Queen’s University, Kingston, Canada
*Corresponding author, e-mail ricardo.rb.academico@gmail.com
KEYWORDS
Tropical Andean reservoir, internal waves, density currents, vertical mixing.
EXTENDED ABSTRACT
Introduction
Observations have revealed density currents are the main precursor of seasonal water
column temperature changes in tropical Andean reservoirs. In turn, the water column changes
modify the mean and turbulent characteristics of the density currents and basin-scale internal
waves. An understanding of turbulence mechanisms is crucial for a good comprehension of
water quality in lakes and reservoirs. The influence of density currents and basin-scale
internal waves on mixing at different time scales has not been investigated in tropical Andean
reservoirs. Motivated by this, we investigated the vertical mixing in Porce II reservoir, an
elongated Colombian reservoir located at 06°47’00”N. Porce II is 9.8 km long with a
maximum depth of 92 m and it is fed by Porce River.
Materials and methods
Microstructure profiles of temperature and turbidity (~1mm resolution) and velocity (0.8
m bin-size) were measured in four short field campaigns carried out during dry and wet
seasons between 2015 and 2016, when a marked difference was found in the vertical
temperature stratification between seasons (Figure 1). The measurements were done in
periods of different external forcing: before and after flood events and in weak and strong
winds (Figure 1). We estimated different mixing indicators such as Reb (Buoyancy Reynolds
number) and Kz (vertical eddy diffusivity) following Bouffard and Boegman (2013)
(hereafter BB), LT (Thorpe scale) and the FrT-ReT diagram (turbulent Froude-Reynolds
numbers) proposed by Ivey and Imberger (1991) (hereafter I&I). The location where the data
collection was done is not directly influenced by selective withdrawal, neither by differential
cooling. During the measurement, negative atmospheric heat fluxes were not observed, so
there was no convective mixing in the surface layer.
Figure 1. Evolution of wind, Porce inflow (a-d) and temperature in 48 m depth, at the measurement station (e-h).
Gray rectangles (a-d) and white dashed-rectangles (e-h) show the times of measurements. Black countours are
each 0.5°C. The white contour marks 24°C. Triangles mark the period of profiles in Figure 2.
Results and discussion
The observations revealed distinctive transport processes, being the basin-scale
internal waves and density currents the most predominant. The amplitude of internal
Physical Processes in Natural Waters 2017
2
oscillations was larger during the dry periods, when the reservoir was more stratified but also
the winds were the strongest (Figure 1a-d). The internal wave field had different structure
according to the season, with V1H1 and V2H2 modes in transition and dry seasons
respectively, but an indistinct mode in the wet season (Posada-Bedoya et al, 2017). Density
currents behaved always as intrusive plumes, being shallower and less turbid in the dry season
(Figure 2a-b), although in a flood event during the dry season (13/Apr) the plume was deeper,
and changed the vertical temperature structure (Figure 1h). During basin-scale internal waves
the generated turbulence was strongly stratified turbulence (Figure 3a, most points near the
limits of Reb and FRϒ around 8 and 3.9 respectively, with FrT < 1), a characteristic of high
stratification, low dissipation with anisotropic, and possibly youthful, turbulence (the ratio
LT/LO, with LO the Osmidov scale, increases above 1 as the non-dimensional parameter LTν =
LT(N/ν)0.5 is large (Mater et al, 2015), Figure 3b). There was even no turbulence, as the
regime for some patches was laminar (Reb (FRϒ) < 8 (3.9), ReT < 15 and Kz~10-7m2/s). The
flood events generated, in general, isotropic-high turbulence (Reb~400 and higher, FrT~1 for
large ReT and LT/LO~1 as LTν increases) with larger Thorpe scales (Figure 3c). As expected,
the turbulence in the surface layer was also high and isotropic while the winds were strong,
common behaviour in the dry season.
Figure 2. Bin and time averaged profiles (1 m bin per day) of a) turbidity, c) velocity (no velocity April 11), d)
Reb (green vertical lines define turbulent regimes according to BB) and e) vertical eddy diffusivity.
Figure 3. a) FrT-ReT diagram and limits by following I&I, b) The ratio LT/LO as a function of LTν (Mater et al,
2015) and c) Reb as function of LT. Gray dashed lines in a) and c) mark the Reb regimes and in b) the ratio
LT/LO =1. In the figures the rectangles mark the plume zone in April 13 (below of 23m), diamonds the mixing
zone (above of 4m) and the circles the remaining data. Black cross-dashed line in c) is a running mean of Reb.
REFERENCES
Bouffard, D., and Boegman, L. (2013). A Diapycnal Diffusivity Model for Stratified Environmental Flows.
Dynamics of Atmospheres and Oceans, 61–62, 14–34.
Ivey, G. N., & Imberger, J. (1991). On the Nature of Turbulence in a Stratified Fluid. Part I: The Energetics of
Mixing. Journal of Physical Oceanography, 21(5), 650–658.
Mater, B. D., Venayagamoorthy, S. K., St. Laurent, L., & Moum, J. N. (2015). Biases in Thorpe-Scale Estimates
of Turbulence Dissipation. Part I: Assessments from Large-Scale Overturns in Oceanographic Data. Journal
of Physical Oceanography, 45(10), 2497–2521.
Posada-Bedoya, A., Gómez-Giraldo, A., Román-Botero, R. Effect of density currents on the seasonal evolution
of basin-scale internal waves in a Tropical Andean reservoir. In preparation, also in PPNW2017.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Distribution of sea water natural constituents on shelves of Black
Sea and Brazilian coast obtained remotely from board a ship
Vera Rostovtseva, Igor Goncharenko and Boris Konovalov
P. P. Shirshov Institute og Oceanology RAS, Moscow, Russia
*Corresponding author, e-mail vera@ocean.ru
KEYWORDS
Ecology of coastal waters, passive optical remote sensing, natural water constituents concentration.
EXTENDED ABSTRACT
Introduction
For studying of the ecological state of shelf waters it is often necessary to obtain data
from a shipborne or an airborne measuring complex operating remotely [Mouw, 2015]. We
used a three-channel passive optical spectrophotometer that enables us to get the sea
reflectance coefficient spectra from board a moving ship. The data of the measurements were
processed then according to our original method, which is based on the intrinsic properties of
the pure water absorption spectrum – water absorption step method (WASM) [Rostovtseva,
2016]. It gives us the possibility to obtain estimates of the absorption spectra of the sea waters
under exploration. The retrieved spectra in its turn were the source of information about water
constituents concentration.
Materials and methods
The measurements were carried out with a semiautomatic measurement complex EMMA
(Ecological Monitoring of Marine Aquatories) operating from board a ship at 1 Hz frequency.
It includes three hyperspectral photometers, the data from which are processed by special
algorithm on base of WASM. In natural waters we can get estimates of phytoplankton
pigments, “yellow substance” and suspended matter concentrations. An example of EMMA
operating from board a small ship and its main characteristics are given in Fig. 1.
EMMA: 3 STS-VIS
+ 2 fiber with lenses
+ 1 fiber with cosine
corrector
STS-VIS:
Dimensions: 40mm*42mm*24mm
Weight: 60g
Detector: ELIS1024
Wavelength range: 350-800 nm
Integration time: 10μs – 10s
Signal-to-noise ratio: >1500:1
(maximal signal)
Dark noise: < 3counts rms
Slit: 50 μm
Optical resolution: 3 nm
Stray light: <0.25% at 590 nm
Fiber optic connector: SMA905
Figure 1. Monitoring of the sea water from board a moving ship with a semiautomatic
measurement complex EMMA
2
Physical Processes in Natural Waters 2017
Results and discussion
The data from the new semiautomatic complex EMMA obtained during the operative
monitoring of coastal waters from board a moving vessel are given for two shelf regions with
different types of sea waters: for the Black Sea coastal waters of oligotrophic and mesotrophic
types as well as for Brazilian coastal waters at the Rio-Grande river mouth of eutrophic type.
From every three measured at the same time spectra for the sea, the sky and the overall water
surface illumination we calculated the sea reflectance coefficient spectra (this value being
dimensionless is proportional to dimensional reflectance). Some typical sea reflectance
coefficient spectra for both regions under investigation are given in Fig. 2.
Sea reflectance coefficient spectra for the
Brazilian coastal waters at the Rio-Grande mouth
0,1
0,09
0,08
0,07
0,06
Sea reflectance coefficient spectra
for the Black Sea
0,05
0,04
0,06
0,03
0,02
0,05
0,01
0,04
350
0
400
450
500
550
600
650
700
750
800
wavelength, nm
0,03
0,02
0,01
350
0
400
450
500
550
600
650
700
750
800
wavelength, nm
Figure 2. Sea reflectance coefficient spectra measured in oligotrophic, mesotrophic (the Black
Sea) and eutrophic (the Brazilian coast) water types
The average difference in magnitude between the spectra from these regions (2 – 5 times)
is not as big as the difference in constituents concentration (about 2 orders of magnitude). So
we considered mostly the difference in shapes. Using such spectra with the help of WASM
we estimated first the water absorption and then the concentrations of the sea water natural
constituents such as suspended matter and coloured organic matter. Distributions of these sea
water natural constituents over the shelves of the Black Sea and Brazilian coast were mapped.
The obtained distributions are in good coincidence with the results of measurements in water
samples and with the satellite data.
This work has been supported by Russian Scientific Fund Project N 14-50-00095 and by
Grant RFMEFI61315X0050 of Russian Education and Science Ministry (in the part of
measurements in Brazil).
REFERENCES
Mouw C. B., S. Greb, D. Aurin et al. (2015), Aquatic color radiometry remote sensing of coastal and inland
waters: Challenges and recommendations for future satellite missions, Remote Sensing of Environment,
160, 15-30.
Rostovtseva V. V. (2016), Method for sea water absorption spectra estimation on the basis of shipboard passive
remote sensing data and pure sea water properties, Atmospheric and Oceanic Optics, 29(2), 162-170.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
On the feasibility of kinetic energy production by Daphnia diel
vertical migration
S. Simoncelli1*, S. J. Thackeray2 and D. J. Wain1
1
Department of Architecture and Civil Environmental,
University of Bath, Bath, United Kingdom
2
Centre for Ecology and Hydrology, Lancaster Environment Centre
Lancaster, United Kingdom
*Corresponding author, e-mail s.simoncelli@bath.ac.uk
KEYWORDS
Biomixing; Daphnia; turbulence; eddy diffusivity; zooplankton.
EXTENDED ABSTRACT
Introduction
Biomixing refers to the contribution of living organisms towards the mixing of waters
in oceans and lakes. Our project focuses on the stirring generated by Daphnia spp. in a small
lake. This very common zooplankton species is engaged in a vertical migration (DVM) at
sunset, with many organisms crossing the thermocline despite the density stratification. During
the ascent, they may create hydrodynamic disturbances in the lake interior where the
stratification usually suppresses the vertical diffusion.
Experimental measurements in an unstratified tank by Wilhelmus & Dabiri (2014) show
that zooplankton can impart kinetic energy at length scales bigger than organism’s size of few
mm through collective motions. Noss & Lorke (2014) measured instead the eddy diffusivity
KV from Daphnia induced migration in a stratified water tank, suggesting that the zooplankton
species generates no mixing, with KV=10-9 m2 s-1. Very recently, DNS by Wang & Ardekani
(2015) showed instead a maximum diffusivity of KV=10−6 m2 s-1 for millimetre-sized swimming
organisms, therefore supporting the idea of zooplankton-generated mixing.
Materials and methods
Measurements were conducted on three different days (21 July, 28 July and 18 August
2016), during the summer stratification, in Vobster Quay, a small and deep (40m) quarry with
small wind fetch and steep sides, located in the south-west UK. Zooplankton vertical
concentration was evaluated using a zooplankton 100-μm mesh net, by collecting and analysing
samples in 8 layers of the lake. A bottom-mounted ADCP was also employed to track their
concentration and migration with the measured backscatter strength. Turbulence and mixing
were measured before and during the DVM with the SCAMP, a microstructure temperature
profiler. Dissipation rates of turbulent kinetic energy (TKE) were determined by fitting the
Batchelor spectrum of temperature fluctuations and removing invalid spectra with statistical
criteria by Ruddick et al. (2000).
Results and discussion
Figure 1 shows the time series of TKE dissipation rates ε in the foreground (top colour-bar)
overlapped with the estimated Daphnia’s concentration (bottom grey colour-bar). Data with ε
< 10−9 W kg-1 are not shown because they represent undisturbed water with background
turbulence conditions. Our data do not show evidence of intensified turbulence during the
vertical ascent of Daphnia: turbulent patches do not increase either in frequency or in
dissipation magnitude on any of the three different sampling dates, despite high Daphnia
Physical Processes in Natural Waters 2017
2
concentrations up to 90 org. L-1 during the DVM in the migrating layer. We observed no
correlation between the zooplankton concentration and turbulence. Moreover, turbulent patches
were too sporadic to create mixing, even if they were generated by the migrating zooplankton.
Part of the observed dissipations might have been associated with penetrative convection or
wind-driven motions but in any case, turbulence was not effectively enhanced enough to prevail
over other shear-generated instabilities.
This seems to suggest that migrating Daphnia do not globally affect the mixing regime.
However, microstructure profilers based on temperature fluctuations measurements may not be
suitable for turbulence sampling if the generated eddy diffusivity is less than 10−7 m2 s-1.
Figure 1: TKE dissipation rates ε in the foreground (top colour-bar) overlapped with the estimated Daphnia’s
concentration (bottom grey colour-bar). Blue dashed line shows the sunset time when DVM begins.
REFERENCES
Noss, C., & Lorke, A. (2014). Direct observation of biomixing by vertically migrating zooplankton. Limnology
and Oceanography, 59(3), 724–732.
Ruddick, B., Anis, A., & Thompson, K. (2000). Maximum Likelihood Spectral Fitting: The Batchelor Spectrum.
Journal of Atmospheric and Oceanic Technology, 17(11), 1541–1555.
Wang, S., & Ardekani, A. M. (2015). Biogenic mixing induced by intermediate Reynolds number swimming in
stratified fluids. Scientific Reports, 5, 17448.
Wilhelmus, M. M., & Dabiri, J. O. (2014). Observations of large-scale fluid transport by laser-guided plankton
aggregations. Physics of Fluids, 26(10), 101302.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Measurement of
Greenhouse Gas Emissions from Reservoirs
U. Spank1, M. Hehn1, P. S. Keller2, C. Bernhofer1 and M. Koschorreck2
1
Chair of Meteorology, Department of Hydrosciences, Faculty of Environmental Sciences,
Technische Universität Dresden, Germany
2
Department of Lake Research, Helmholtz Centre for Environmental Research GmbH – UFZ,
Germany
*Corresponding author, e-mail uwe.spank@tu-dresden.de
KEYWORDS
Eddy covariance; methane and carbon dioxide emissions; ebullition; temporal and spatial variability;
meteorological impact.
EXTENDED ABSTRACT
Introduction
The emissions of carbon dioxide (CO2) and methane (CH4) from inland waters are an
important source in the global greenhouse gas (GHG) balance (Bastviken et al., 2011;
Raymond et al., 2013). Reservoirs are particular hot spots of GHG emissions (Deemer et al.
2016). The GHG emissions are temporally and spatially highly variable (Zhao et al., 2013).
Currently, there is not much known about the actual flux rates from temperate reservoirs, and
the processes involved are not completely understood. In our recently started project TregaTa,
we aim to quantify the GHG emissions from two German reservoirs. The central topic is the
understanding of the regulation of CO2 and CH4 emissions and, especially, to comprehend
effects of altering water levels, trophic state and meteorological drivers. We would like to test
three major hypotheses:
(1) Short term events contribute significantly to the overall balance.
(2) Temporal patterns of CO2 and CH4 emissions depend on the trophic state of the
reservoir and are complexly overlaid by atmospheric effects.
(3) The spatial distribution of the CO2 and CH4 fluxes depends differently on internal
(e.g., hydro-chemical parameters and water depth) and external interacting factors (e.g., wind,
air pressure, radiation and energy balance).
Our project is placed in the nexus between limnology, hydrology and boundary layer
meteorology. We will investigate two different reservoirs – the mesotrophic Rappbode
reservoir in the Harz Mountains and the eutrophic reservoir Bautzen in Lusatia. Temporal
patterns of CO2 and CH4 emissions will be quantified by a combination of micrometeorological and water-side measurements. Central to the project is a floating Eddy
Covariance (EC) measurement system. The conflation of EC flux measurements, in situ
concentration measurements of the water body, and meteorological basic data allows the
determination of the physical gas transfer coefficient. Additionally, the spatial variability of
GHG emissions will be analysed by floating chamber measurements and ebullition funnels
(bubble traps). The measurements of emission rates are accompanied by the analysis of
sediment and hydro-chemical parameters, e.g., pH and concentrations of O2, CO2 and CH4 as
well as by continuous measurements of energy balance, radiation budget and classical
meteorological variables. Proper modelling approaches will be used for generalisation and
regionalisation of measurements and project results.
Physical Processes in Natural Waters 2015
2
Materials and methods
A floating measurement station has been established on Rappbode reservoir in March
2017 to measure greenhouse gas (GHG) emissions and energy exchange between water
surface and atmosphere via eddy covariance (EC) technique. EC measurement are the stateof-the-art method to monitor carbon dioxide flux exchange (F CO2) and methane emissions
(FCH4) as well as sensible (H) and latent heat fluxes (LE). However, EC measurements on
floating platforms are not a standard task but imply several technical and scientific challenges.
Notwithstanding that the concept of a floating platform is an elegant opportunity for EC
measurements which are largely unaffected by surrounding land surfaces. Our prototype of a
floating platform is easy to assemble and provides a proper basis for ‘offshore’ observations.
The EC measurement system device consists of an ultrasonic anemometer (Campbell
CSAT3, Campbell Scientific Ltd., UK) for measurements of 3-dimensonal wind velocity, a
closed-path gas analysers, LI-COR LI-7000 (LI-COR Biosciences, USA) for high frequency
concentration measurements of carbon dioxide and water vapour in air, an open-path gas
analyser, LI-COR LI-7700 (LI-COR Biosciences, USA) for high frequency measurements of
methane concentration in air, a logger unit (Campbell CR6, Campbell Scientific Ltd., UK)
and a high frequency inclinometer (SCA114, a.b.jödden GmbH, Germany) to monitor
platform movement. The EC system is operated with 10 Hz, i.e., data of 3-dimensonal wind
speed, air temperature, platform’s roll and pitch angle as well as air concentrations of
methane, carbon dioxide and water vapour are measured and logged 10-times per second.
Based on high frequency raw data, exchange rates, i.e., H, LE, F CO2 and FCH4, are
processed according to standards of EUROFLUX community (Aubinet et al., 2012) and usage
of the software package EdiRe (The University of Edinburgh, 2007) in half-hourly resolution.
Outliers and data beyond absolute limits are eliminated, covariances and raw fluxes are
calculated based on half-hourly block averages, and time lags between wind speed and gas
concentration measurements are removed applying covariance maximisation. The axis
rotation for tilt correction is applied using double rotation method as described by (Wilczak et
al., 2001). However, standard proceeding of tilt correction is extended by algorisms correcting
actual roll and pitch angles of the platform measured by the attached inclinometer.
Furthermore, flux processing includes algorisms for spectral attenuation (Spank and
Bernhofer, 2008), sonic temperature correction (Schotanus et al., 1983) and density
fluctuations (Webb, 1982). The quality of processed fluxes is assessed according to Foken
and Wichura (1996)
Besides turbulent fluxes, radiation balance, i.e., all four components of radiation balance
(Campbell CNR1, Campbell Scientific Ltd., UK; installed 1m above water surface) and water
temperature in different depths are measured for assessment of energy balance and energy
exchange of water body. Sensors (LI-COR LI-190SZ, LI-COR Biosciences, USA) to monitor
incoming and reflected photosynthetic active radiation are installed 1 m above the water
surface. Photosynthetic active radiation is also measured in water depths of 0.25 and 4.0 m
(LI-COR LI-193, LI-COR Biosciences, USA) to estimate penetration depth of solar radiation.
In addition, air temperature and air humidity (HMP 45, Vaisala, Finland) are logged.
The micro-meteorological and hydro-physical measurements are supplemented by
continuous hydro-chemical observations of oxygen (O 2), carbon dioxide and methane as well
as special measurement campaigns (executed twice per month) investigating spatial
variability of CO2 and CH4 emissions via chamber measurements. Bubble traps are installed
at several points in the reservoir to investigate ebullition. The oxygen sensors (D-Opto, ZebraTech Ltd, New Zealand) are installed in a depth of 0.25 m and 4 m. The sensors for
continuous measurements of dissolved CO2 (CONTROS HydroC CO2, Kongsberg Maritime
Contros GmbH, Germany) and CH4 (METS methane sensor, Franatech GmbH, Germany) are
installed shallowly below the water surface in a depth of 25 cm. Thus, fluxes of CO2 and CH4,
Spank et al.
3
derived from EC measurement, can be cross checked with fluxes that are classically estimated
from concentration differences between water and air as well as with floating chamber
measurements
Results and discussion
At the workshop, we will present a complete data set of greenhouse gas emissions and
energy exchange form the transition of spring mixing to summer stagnation. Special focus of
the presentation will be on the effect of meteorological ‘extremes’ such as storm events on
emissions of CO2 and CH4. At the same time, we want to provide an honest report on specific
technical challenges and hurdles as well as an initial assessment of the feasibility of ‘offshore’
EC measurements. Figures 1 and 2 show the complete instrumented station and the position
of the station (51° 44.2726’N, 10° 53.4270’E) on the Rappbode reservoir. It should be noted
that the distance to the nearest bank is more than 150m and to the dam 270m.
Figure 1: Floating micro-meteorological and hydro-chemical measurement station on
Rappbode reservoir in Harz Mountains. The poles and wires are for lighning protection.
Physical Processes in Natural Waters 2015
4
Figure 2: Location of Rappbode reservoir in Germany (upper left) and position of the
measurement station on the Rappbode reservoir (right). Also shown bathymetric characteristic
of reservoir
Acknowledgements
This project (TregaTa) is financed and supported by the German Science Foundation
(Deutsche Forschungsgemeinschaft, DFG; SP 1570/1-1). We especially thank the
Talsperrenbetrieb Sachsen-Anhalt for their kind support and cooperation. Special thanks go
also to Udo Postel, Martin Wieprecht, Uwe Eichelmann and Heiko Prasse for their technical
assistance.
REFERENCES
Aubinet, M., Vesala, T., and Papale, D. (2012). Eddy covariance : a practical guide to measurement and data
analysis (Dordrecht; New York: Springer).
Bastviken, D., Tranvik, L.J., Downing, J.A., Crill, P.M., and Enrich-Prast, A. (2011). Freshwater Methane
Emissions Offset the Continental Carbon Sink. Science 331, 50–50.
Deemer, B.R., Harrison, J.A., Li, S., Beaulieu, J.J., DelSontro, T., Barros, N., Bezerra-Neto, J.F., Powers, S.M.,
dos Santos, M.A., Vonk, J.A. (2016). Greenhouse Gas Emissions from Reservoir Water Surfaces: A New
Global Synthesis. Bioscience 66, 949-964.
Foken, T., and Wichura, B. (1996). Tools for quality assessment of surface-based flux measurements. Agric.
For. Meteorol. 78, 83–105.
Raymond, P.A., Hartmann, J., Lauerwald, R., Sobek, S., McDonald, C., Hoover, M., Butman, D., Striegl, R.,
Mayorga, E., Humborg, C., et al. (2013). Global carbon dioxide emissions from inland waters. Nature 503,
355–359.
Schotanus, P., Nieuwstadt, F.T.M., and Bruin, H.A.R.D. (1983). Temperature measurement with a sonic
anemometer and its application to heat and moisture fluxes. Bound.-Layer Meteorol. 26, 81–93.
Spank, U., and Bernhofer, C. (2008). Another simple method of spectral correction to obtain robust eddycovariance results. Bound.-Layer Meteorol. 128, 403–422.
The University of Edinburgh [School of GeoSciences, Institute of Atmospheric and Environmental Science],
2007: EdiRe.- published online: http://www.geos.ed.ac.uk/homes/jbm/micromet/EdiRe/ (accessed at 08.08.
2017).
Webb, E.K. (1982). On the correction of flux measurements for effects of heat and water vapour transfer.
Bound.-Layer Meteorol. 23, 251–254.
Spank et al.
5
Wilczak, J.M., Oncley, S.P., and Stage, S.A. (2001). Sonic Anemometer Tilt Correction Algorithms. Bound.Layer Meteorol. 99, 127–150.
Zhao, Y., Wu, B.F., and Zeng, Y. (2013). Spatial and temporal patterns of greenhouse gas emissions from Three
Gorges Reservoir of China. Biogeosciences 10, 1219–1230.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Earth rotation vs. seiching in lakes:
implications for one-dimensional modelling
V. Stepanenko1,2*
1
Research Computing Center,
Moscow State University, Moscow, Russia
2
Faculty of Geography,
Moscow State University, Moscow, Russia
*Corresponding author, e-mail stepanen@srcc.msu.ru
Keywords
Lakes; seiches; Coriolis force; modelling; mixed-layer dynamics.
EXTENDED Abstract
Introduction
Most 1D lake models developed so far are based on equations for heat and momenutm of boundary-layer
theory taking into account Coriolis force, where horizontal lake sizes are not involved. More sophisticated 1D
models incorporate area-depth dependence introducing exchange of scalars and momentum between water body
and sloping bottom and explicitly specifying the water volume over which momentum, heat and gases are
distributed. These models, however, still lack the gross features of lake dynamics, such as 3D mass distribution,
surface level variations and related horizontal pressured gradients altogether caused by presence of impermeable
boundaries. In lakes, all these features are interconnected by mechanism of gravitational barotropic and
baroclinic waves (seiches), affected by Earth rotation. The linear wave theory tells us that Coriolis force
becomes negligible compared to pressure gradient when the lake is much smaller than the internal Rossby
deformation radius, which is several kilometers in summer in midlatitudes. Thus, for such lakes, models may
neglect rotation but should necessarily account for barotropic and baroclinic horizontal pressure gradient.
However, conventional 1D lake models include Coriolis acceleration, but not horizontal pressure gradient.
Moreover, numerous applications of such models to small lakes demonstrate they are still capable of realistically
reproduce the vertical mixing. This poses the questions, addressed in current study:
– what is the reason for apparent realistic vertical mixing in conventional 1D lake models, whereas the
simulated lake dynamics is far from that in enclosed water bodies?
– what are the ways to parameterize lake seiching motions in 1D models?
Materials and methods
We propose a mathematical approach to include seiche motions into 1D lake model. It combines traditional
1D equations for simulating continuous vertical profiles of momentum, temperature and tracers, with multilayer
model of seiche dynamics, operating with step-wise density profile schematization (e.g. Münnich et al.,
1992). The key hypothesis making it possible is that only 1-st horizontal seiche mode is considered, justified by
a bulk of empirical evidence indicating that this mode contains a major part of internal wave energy (e.g., Horn
et al., 1986; Kirillin et al., 2015).
Consider a stratified basin of uniform depth H with horizontal dimensions Lx and Ly. Assume, that the
vertical density profile may be approximated by N homogeneous layers, of mean thicknesses and densities Hi
and ρi, respectively. Then, under moderate wind forcing the mean velocity components and thickness deviations
hi’ for all layers of the frictionless fluid are governed by a linearised multi-layer set of shallow water equations,
linked by hydrostatic approximation. On the other hand, the momentum equations for these layers may be
derived from the momentum equations of 1D model as well by vertical averaging, providing the same equations
but with extra inter-layer friction terms. As friction reduces hi’, linearised equation for hi’ (continuity equation)
from multi-layer system holds when 1D momentum equations are used to describe average velocities in discrete
layers. This enables to involve hi’-equation to close the horizontal pressure gradient terms in 1D momentum
equations, keeping this gradient constant in each i-th layer. The continuity equation is 2D in horizontal, so the
predefined horizontal structure for hi’ and velocities has to be assumed in order to fit the 1D framework. Hence,
we assume 1-st horizontal mode in both horizontal directions for these variables. The resulting momentum
equations of 1D model take the form:
Physical Processes in Natural Waters 2015
2
where zj, zj+1 are the boundaries of j-th layer, u and v in last two equations (that are derived from continuity
equation) are averaged over j-th layer. The left-hand sides of first two equations represent the traditional 1D
momentum equation terms, while the right-hand sides stand for the pressure gradient force. Hence, the pressure
gradient force is represented as piecewise-constant with depth. The seiche parameterization described above has
been introduced into the 1D LAKE model (Stepanenko et al., 2016).
Results and discussion
The model described above was run in the idealized Kato-Phillips experiment setup, where the wind-driven
mixed layer deepens into the stratified water basin. It is shown that for the small horizontal size of a lake, seiches
are the main contributor to hampering the mixed-layer deepening, while for the lake size much exceeding the
Rossby deformation radius, Coriolis force effect dominates. The side effect of seiches in a small basin is
production of shear-driven turbulence near the bottom, that cannot be reproduced by conventional 1D models.
REFERENCES
Kirillin, G., Lorang, M. S., Lippmann, T. C., Gotschalk, C. C., & Schimmelpfennig, S. (2015). Surface seiches
in Flathead Lake. Hydrol. Earth Syst. Sci, 19, 2605–2615. http://doi.org/10.5194/hess-19-2605-2015
Horn, W., Mortimer, C. H., & Schwab, D. J. (1986). Wind-induced internal seiches in Lake Zurich observed and
modeled. Limnology and Oceanography, 31(6), 1232–1254. http://doi.org/10.4319/lo.1986.31.6.1232
Münnich, M., Wüest, A. & Imboden, D.M., 1992. Observations of the second vertical mode of the internal
seiche in an alpine lake. Limnology and Oceanography, 37(8), pp.1705–1719.
Stepanenko, V., Mammarella, I., Ojala, A., Miettinen, H., Lykosov, V., & Vesala, T. (2016). LAKE 2.0: a model
for temperature, methane, carbon dioxide and oxygen dynamics in lakes. Geoscientific Model
Development, 9(5), 1977–2006. http://doi.org/10.5194/gmd-9-1977-2016
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Coupled methane and oxygen dynamics during distinct periods of
thermal stability in a small Swiss lake (Soppensee)
D. Vachon1*, D. Donis1, T. Langenegger1, S. Flury1,2 and D. F. McGinnis1
1
Aquatic Physics Group, Department F.-A. Forel for environmental and aquatic sciences
(DEFSE), Faculty of Science, University of Geneva, Geneva, Switzerland
2
Stream Biofilm and Ecosystem Research Laboratory, Ecole Polytechnique Fédérale de
Lausanne, Lausanne, Switzerland
*Corresponding author, e-mail dominic.vachon@unige.ch
KEYWORDS
eutrophic lake; methane dynamics; greenhouse gases emission; thermal stratification; methane bubbling.
EXTENDED ABSTRACT
Introduction
Methane (CH4) is a potent greenhouse gas and its emissions from lakes are significant at
the global scale. Lake CH4 emissions to the atmosphere are the net results of production and
removal processes. Unless CH4 is directly emitted through ebullition (McGinnis et al. 2006), it
is often assumed that most CH4 is oxidized before reaching the atmosphere (Bastviken et al.
2008). Physical transport is thus the key to link CH4 production/removal and efflux, as it will
dictate if produced CH4 will be oxidized or emitted. Understanding the complex interaction
between CH4 production/removal and lake hydrodynamics is central quantify CH4 dynamics
under future environmental changes.
Here we resolve the vertical fluxes of CH4 during different stratification periods using
profiles and vertical diffusivities (Kz). We then evaluate the production and consumption rates
by a mass balance approach and compare with laboratory incubations.
Materials and methods
Soppensee, a small (26 ha and maximum depth of 27 m) eutrophic kettle lake situated in
the Canton Lucerne, Switzerland, was sampled monthly from April to October 2016 for CH 4
diffusive fluxes and profiles. A temperature chain was deployed at the deepest point of the lake
to calculate vertical diffusivity (Kz) using the heat budget method.
Monthly water column CTD/O2 and dissolved CH4 (with stable isotopes) profiles were
performed at the deepest point of the lake. Diffusive flux with the atmosphere was measured
using a floating chamber connected in a closed loop with a portable greenhouse gas analyser.
Bubble-mediated CH4 fluxes were estimated using inverted funnels. Vertical diffusive gas (CH4
and O2) fluxes at depth between epilimnion/metalimnion and metalimnion/hypolimnion were
estimated using Fick’s first law of diffusion. We calculated the CH4 mass budget for each layer
of the lake (epi-, meta- and hypolimnion) for each stratification periods (warming, stable and
cooling). Sediments cores were taken to evaluate CH 4 production and diffusive flux. Methane
oxidation was assessed using lake water in situ incubations.
Results and discussion
In Soppensee, O2 saturation varied in the surface layer from 80%, in spring to 230% in
July-August during the strongest stratification period. Bottom waters (10-25m) remained
anoxic until November (Figure 1a). In the epilimnion, dissolved CH4 concentrations were
2
Physical Processes in Natural Waters 2017
always oversaturated, often around 200-300 times the atmosphere saturation, but ranging from
3 to up to 1000 times) (Figure 1b).
a
b
c
Figure 1. Water column profiles of O2 saturation (%), CH4 concentration (Pmol L-1) and CH4 isotopic
signature G13C-CH4 (‰). Grey area represent the metalimnion.
As CH4 diffusion at the epi-/metalimnion interface was nearly zero until fall mixing,
surface fluxes were almost completely independent of processes occurring in the deeper layers.
The net CH4 production rates derived from the mass balance were constant during the stratified
period (~0.34 mmol m-3 d-1). This suggests that summer diffusive fluxes were entirely sustained
by local processes, either coming from the littoral environments (Bastviken et al. 2008), from
bottom water via non-diffusive processes like CH4 bubble re-dissolution (McGinnis et al.
2006), zooplankton-mediated transport (McGinnis et al. 2017) or by in situ production (Donis
et al. in review).
The metalimnion acts as a “CH4 trap”, where it is actively oxidized. Between June and
October, a CH4 minimum peak concentration remained ~0.3 Pmol L-1. Most of the oxidation
activity occurred at the base of the metalimnion, as indicated by enriched 13C-CH4 values
(Figure 1c). Rates of oxidation derived from the mass balance agreed well with incubation rates
(~ -0.3 Pmol L-1 d-1). We calculated that CH4 oxidation accounted for about 30% and 18% of
the net O2 consumption in the metalimnion during the warming and stable periods, respectively.
In the hypolimnion, dissolved CH4 accumulated due to strong water column stability to
almost 1000 Pmol L-1 (Fig 1). The rate of CH4 accumulation (below 10 m) was an order of
magnitude higher than what Bastviken et al. (2008) found in two oligotrophic lakes, even
though sediment fluxes were comparable. We estimated from the mass balance sediment CH4
production rates of 9.7-13.7 mmol m-2 d-1, in good agreement with the CH4 diffusion and
production rates from incubations.
Our results show that in this small eutrophic lake, summer emissions are sustained and
independent of deep-water methane. The vertical diffusivity strongly modulates O2 and CH4
vertical fluxes and accumulation, which in turns drives the CH4 fate and emission.
Implementing robust lake hydrodynamics is necessary to refine constituent budgets leading to
robust insights on their fates.
REFERENCES
Bastviken, D., J. J. Cole, M. L. Pace, and M. C. Van de-Bogert, (2008), Fates of methane from different lake
habitats: Connecting whole-lake budgets and CH4 emissions. J. Geophys. Res. Biogeosciences 113: 1–13.
McGinnis, D. F., J. Greinert, Y. Artemov, S. E. Beaubien, and A. Wüest. (2006), Fate of rising methane bubbles
in stratified waters: How much methane reaches the atmosphere? J. Geophys. Res. (C Oceans) 111(C9): 15.
McGinnis, D. F., S. Flury, K. W. Tang, and H.-P. Grossart, (2017), Porewater methane transport within the gas
vesicles of diurnally migrating Chaoborus spp.: An energetic advantage, Sci. Rep. 7: 44478.
Donis, D., S. Flury, A. Stöckli, J. Spangenberg, Vachon, D. and McGinnis, D. F. (in review), Large methane
emissions from the surface layer of a mesotrophic lake: A full-scale evaluation of the “methane paradox”.
Nat. Comm.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Radiatively-driven convection in a small ice-covered lake:
Dynamics of velocities and energy dissipation
S. Volkov*, G. Zdorovennova, R. Zdorovennov, T. Efremova, N. Palshin, A. Terzhevik and
S. Bogdanov
Northern Water Problems Institute, Russian Academy of Sciences, Petrozavodsk, Russia
*Corresponding author, e-mail taranarmo@gmail.com
KEYWORDS
Ice-covered lakes; convection; energy dissipation; convective cells dynamic.
Introduction
Radiatively driven convection in ice-covered lakes is a unique natural phenomenon
suitable for studying the fundamentals of turbulence onset and development in a stratified
fluid. The convective mixing during late spring, being the most energetic event during the ice
season, governs the dynamics of important chemical and biological processes. A number of
papers are devoted to studying the mixed layer properties and penetrative convection
dynamics, but some topics still remain challenging, including in particular 2D-3D interplay,
turbulent transport specifics, intermittency effects. These challenges are only stressed by some
LES and direct numerical simulations. More detailed experimental data is necessary for
further progress.
Materials and methods
For Lake Vendyurskoe, a typical shallow (max depth 13 m) boreal lake, located at
southern part of Russian Karelia, the first observations of convection under the ice were taken
in spring 1994 and 1995. During 2000’s seasons the structure and dynamics of water
temperature have been studied in detail by analyzing the observational data [Mironov et al.,
2002]. A new stage of the research began in April 2016, when the Aquadopp HR-profiler was
used for measuring all three velocity components. The scanned region included the 2-m thick
layer under the ice. Spatial (vertical) resolution of 5 cm was chosen with records one minute
apart taking into account the values of rd = (η3/ε)1/4 ~ 3 mm and td = (η/ε)1/2 ~ (10-30) s of
Kolmogorov scales.
Results and discussion
The prior attention is paid to mean velocity and pulsations profiles and its diurnal
dynamics, along with the energy dissipation rate H. The Temperature Microstructure Method
based on fitting the experimental series with Batchelor spectra was explored in [Jonas et. al.,
2003] for evaluating H. One of the alternative ways to derive the value of H is based on the
analysis of 2-point structure functions. With available experimental data only the following
types of longitudinal and transverse structure functions are available
DNN ( z ) (u ( z 0 'z ) u ( z 0 )) 2
DLL ( z )
( w( z 0 'z ) w( z 0 )) 2
Here z0 is the reference point; u and w are horizontal and vertical velocity fluctuations,
respectively; < > denotes the averaging with time.
Both functions are calculated for different time intervals and depths z0. In most cases
the presence of the so-called inertial interval was proved. Namely, the classical asymptotic
dependence D = C H2/3 r2/3 valid for L ≫ ' z ≫ rd, is detected (L is the thickness of the layer).
For both structure functions the r2/3 – interval is clearly manifested in daytime within one
Physical Processes in Natural Waters 2017
2
decade and even more: 0.05 m < ' z < (0.5-1) m as presented in the Figure. This result can be
regarded as clear evidence of the fully developed turbulence regime.
Figure 1: Transverse structure function evolution during daytime. Apr 10 2016. Left – DNN('z), right –
DNN('z⅔), reference point – 1 m.
As for estimation of H, it was straightforward, through identification of inertial interval
and applying least squares method for analyzing linear correlation between D and r2/3. The
typical values vary within few units of 10-8 Wt/kg during daytime with mean value close to
4*10-8. The mean error was about 5 %. It must be stressed that calculated values for vertical
and horizontal rates of energy dissipation are sufficiently different. This finding strongly
indicates the anisotropy of pulsations even in the inertial interval. In night series, the inertial
interval is rather hidden, and H estimations are not fully accurate. However, the rough
correlation of H(t) with radiation dynamics is obtained.
Last but not least, the question concerning the direction of energy transfer arises.
Taking into account both the existence of inertial interval and the dynamics of Reynolds stress
tensor components, we can conclude that the transfer is direct (large to small scales) during
daytime.
As for further activity, we plan to concentrate on
- More detailed analysis of correlation between dynamics of solar radiation and H, in
particular: do there exist time lags, radiation thresholds etc.
- The study of the mechanisms of energy supply and anisotropy survival.
- The thorough analysis of high-order (at least – cubic) 2-point structure function in
order to shed an additional light on the specifics of direct energy transfer.
The study was supported by the Russian Foundation for Basic Research (project 1605-00436_a).
REFERENCES
Mironov, D., A. Terzhevik, G. Kirillin, T. Jonas, J. Malm, and D. Farmer (2002), Radiatively driven convection
in ice-covered lakes: Observations, scaling, and a mixed layer model. J. Geophys. Res., 107(C4),
doi:10.1029/2001JC000892.
Jonas, T., A. Terzhevik, D. Mironov, and A. Wüest (2003), Radiatively driven convection in an ice-covered lake
investigated by using temperature microstructure technique, J. Geophys. Res., 108(C6), 3183,
doi:10.1029/2002JC001316.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August 2017
Combining downstream river demands with a sustainable raw
water supply from a drinking water reservoir
M. Weber1*, K. Rinke1 and B. Boehrer1
1
Department of Lake Research, Helmholtz Centre for Environmental Research - UFZ,
Magdeburg, Germany
*Corresponding author, e-mail michael.weber@ufz.de
KEYWORDS
Reservoir management; reservoir modelling; selective withdrawal; thermal stratification; hypolimnetic dissolved
oxygen.
EXTENDED ABSTRACT
Introduction
Dams and reservoirs interrupt the river continuum and affect water temperature,
discharge, sediment transport and nutrient availability downstream. Freshwater resources like
rivers and drinking water reservoirs must be specially protected in times of global change and
population growth. Therefore, the demands of a safe raw water supply and a good ecological
status of the downstream river should be balanced.
Selective withdrawal is a common practice in reservoir management to control
downstream discharge and in-reservoir water quality. While selective epi-/metalimnetic
withdrawal allows to reconstruct the natural downstream temperature, hypolimnetic
withdrawal through the bottom outlet controls the dissolved oxygen (DO) content in the deep
reservoir layers. In this study, we aim at improving the current withdrawal strategy focussing
on three goals:
- restoring the natural temperature and flow discharge downstream,
- securing raw water quality by maximizing hypolimnetic oxygen,
- minimizing the loss of hypolimnetic water.
Because all three goals are inter-linked and could not be altered without a feedback on the
others, we use a numerical reservoir model to link water quantity management with water
quality.
Materials and methods
We set-up the numerical model for the second largest drinking water reservoir in
Germany, the oligotrophic and monomictic Grosse Dhuenn Reservoir (max. volume
81 million m ). For a realistic withdrawal management, we included reservoir management
parameters and operational rules into the source code of the model. Now, the model is able to
determine the best height for selective epi-/metalimnetic withdrawal based on stratification
and the interaction with the bottom outlet withdrawal during runtime. First, we identified the
practicability of a new withdrawal strategy that focuses on the temperature and flow
requirements of the downstream river. Second, we searched in scenario simulations for the
ideal balance of both withdrawal types to avoid low-oxygen conditions in the deep
hypolimnion while keeping discharge temperature natural.
The reservoir model and its source code are freely available for download from GitHub
(https://github.com/AquaticEcoDynamics).
Physical Processes in Natural Waters 2017
2
Results and discussion
We found that a predominant withdrawal from the epi-/metalimnion severely affected
stratification and decreased hypolimnetic DO concentrations. Using a parallel withdrawal
through the bottom outlet kept DO concentrations sufficiently high enough (3-4 mg/L) to
guarantee a safe raw water supply (Fig. 1c). Scenario simulations showed that a smart mixing
of cold hypolimnetic water with warm epi-/metalimnetic water to follow a target downstream
temperature and flow discharge can be achieved without jeopardizing raw water quality.
The model was able to restore the natural temperature downstream as closely as possible
(Fig. 1b, RMSE 1.7 °C for summer period). The average water temperature of the withdrawn
water was 11.7 °C and approximately 6.3 °C higher than with a withdrawal strategy using the
bottom outlet withdrawal only. Our new withdrawal strategy with only four parameters can be
easily integrated into the operational use.
Fig. 1. (a) Contour plot of simulated water temperatures of the Grosse Dhuenn Reservoir assuming improved
withdrawal strategy over the period of 1996-2013; the solid magenta line indicates the autonomously determined
withdrawal height for the pivoted pipe (selective epi-/metalimnetic withdrawal) on basis of upstream river
temperature and flow discharge as a target; the dashed magenta line indicates withdrawal height and time, when
the bottom outlet was activated. (b) Upstream river temperature as target discharge temperature (red solid line)
versus original operated (only bottom outlet, blue solid line) and improved (both withdrawals, black solid line)
discharge water temperature achieved by the model. (c) Simulated dissolved oxygen concentrations in 5 m above
bottom (green solid line) of the improved withdrawal strategy versus the threshold for the model of 4 mg/L (red
solid line).
REFERENCES
Hipsey, M.R., L.C. Bruce and D.P. Hamilton (2014), GLM - General Lake Model: Model overview and user
information. AED Report #26, The University of Western Australia, Perth, Australia.
Weber, M., K. Rinke, M.R. Hipsey and B. Boehrer (2017),Optimizing withdrawal from drinking water reservoirs
to reduce downstream temperature pollution and reservoir hypoxia, J. Environ. Manag.,197, 96-105.
ABSTRACTS
20th International Physical Processes in Natural Waters (PPNW) Workshop, 21-25 August 2017
Flowpath and retention of snowmelt in an ice-covered Arctic lake
A. Cortes*, S. Sadro, and S. MacIntyre
University of California, Santa Barbara, California, USA
*Corresponding author, alicia.cortes@ucsb.edu
ABSTRACT
The extent to which snowmelt flowing into ice-covered lakes spreads horizontally and
mixes vertically influences retention of solutes derived from the landscape. To quantify these
transport processes and retention, we combine time series temperature and specific conductance
measurements in Toolik Lake (Alaska) and its major inflow, with measurements of discharge
and meteorology, and profiles of specific conductance, temperature, fluorescence, chlorophyll a
and dissolved organic carbon (DOC) in spring of 3 years. During early snowmelt, the
concentration of DOC in the stream was 750 µM, twice that in the lake. During slow melt
(discharge (Q) < 4 m3 s-1), the incoming solute-rich intrusion spread lakewide below the ice.
During melt with Q > 6 m3 s-1, the incoming water partially flushed the inlet basin and the more
dilute water flowed over the original intrusion with a preferential flowpath to the outlet.
Penetrative convection was restricted by the increased density gradients from the incoming
plume and initially constrained to shallow mixing zones associated with the step changes in
density. As ice thickness decreased to less than 1 m, heating caused density instabilities at the
base of the intrusions that mixed solutes ~10 m vertically. Near-surface layers enriched with
DOC persisted for over 3 weeks when melt was slow but ~10 days during a rapid melt when
15% of the solutes introduced into the lake were retained. Variability in retention additionally
depends on the extent of mixing within the plume on the preferential flowpath and on the onset
of deep mixing from penetrative convection.
KEYWORDS
Arctic lakes, ice cover, snowmelt inflow, dissolved organic carbon
20th International Physical Processes in Natural Waters (PPNW) workshop
EVALUATING ALGAL-DRIVEN SHIFTS IN COASTAL SEDIMENT-WATER OXYGEN DYNAMICS
Rebecca Ellis, University of Bath/ Plymouth Marine Laboratory
Key biogeochemical processes at the sediment-water interface, including decomposition of
precipitated algal matter and bio-irrigation, have been shown via certain studies to have a
significant influence on sediment-water oxygen (O2) uptake. Yet this seasonal component of
water-column O2 dynamics is frequently overlooked in assessments of O2 budgets. The main
objective of this research is to comprehend in situ dissolved O2 dynamics on seasonal scales. This
study focuses on elucidating and quantifying the sediment-water interface O2 budgets at a coastal
site. In essence, this project provides a comprehensive analysis of the water column O2 budget,
with particular focus on the influence of seasonal algal blooms and sedimentary biochemical
processes off the coast of Plymouth on sediment O2 uptake. An eddy covariance system is used
to assess turbulence accompanied by supporting laboratory experiments evaluating the influence
of bio-irrigation on sediment O2 uptake. All equipment is deployed over one seasonal cycle at the
Western Channel Observatory. L4 is a long-term coastal monitoring site that is easily accessed
from Plymouth Marine Laboratory. The L4 study site is an excellent location due to the extensive
historical datasets for the water column (> 20 years) and benthos (> 50 years). The novel O2 data
obtained via this project will be combined with the pelagic and benthic datasets to gain important
understanding of the O2 dynamics at L4.
Quantifying Spatial Variability in a Deep, sub-Alpine Lake
A.L. Forrest1,2, K. Le1, J. B.T. McInerney1, G. Schladow1,2 and J. Austin3
1
Department of Civil & Environmental Engineering,
University of California – Davis, Davis, CA 95618, USA
2
Tahoe Environmental Research Center,
University of California – Davis, Incline Village, NV 89451, USA
3
Large Lakes Observatory and Department of Physics,
University of Minnesota Duluth, Duluth, MN 55812, USA
*Corresponding author, e-mail alforrest@ucdavis.edu
KEYWORDS
Spatial variability; surface mixing; horizontal spectra; autonomous underwater glider
Lakes of all sizes are known to display heterogeneity in both the horizontal and vertical
directions. The causes range from interactions with boundaries, distributed wind and radiation
patterns and, in larger lakes, rotational effects. This spatial variability in the upper water
column was investigated in Lake Tahoe (CA-NV, USA), a deep (maximum depth of 501 m),
sub-alpine lake located in the Sierra Nevada using an autonomous underwater glider. The glider
was programmed to run repeated dives to 150 m depth at a 26º dive angle along a repeat transect
15 km long across the widest, central region of the basin. A total of 17 transects were run over
11 days, with a typical transect taking 15 hours. As the glider operated continuously, the same
parts of the lake were sampled at different times from transect to transect. Key variables being
measured included temperature, specific conductivity, dissolved oxygen, chlorophyll
fluorescence and optical backscatter in the water column. The sampling period covered a range
of different lake forcing, including major wind-driven upwelling events. Aggregating the
transects allow estimates of the mean and standard deviation to be made for any location within
the single cross-sectional matrix of the lake for the observed time period. In the first instance,
such an exercise provides quantitative estimates of the inherent heterogeneity as a function of
both depth and position across the lake. Horizontal spatial spectra were also derived for fixed
depths of 5 m (in the surface mixed layer), 20 m (at the approximate seasonal thermocline) and
50 m (in the metalimnion) across each transect by assuming a constant interval spacing. The
spectra slope of the derived passive tracer spectrum at these three depths allows estimates of
the mixing heterogeneity at these three depths to be derived. Understanding this natural
variability of the physical processes within this large lake system is critical for interpreting data
from ongoing lake monitoring programs and for customizing future experimental designs.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä, Finland, 21-25 August, 2017
Remobilization and transport of particles in the nearshore zone of
Lake Constance
H. Hofmann1 and P. Dissanayake1
1
Environmental Physics Group, Limnological Institute, University of Konstanz, Mainaustr.
252, D-78465 Konstanz, Germany
*Corresponding author, e-mail hilmar.hofmann@uni-konstanz.de
Oral presentation
ABSTRACT
Many lake shores experience erosion due to modifications in shore morphology, e.g. harbors
and piers, long-term changes in water level and storm events. It is of upmost importance to
understand the governing processes of erosion, e.g. the interaction between the nearshore
hydrodynamics, sediment remobilization, and transport. Sediment dynamics in the nearshore
zone of lakes are strongly affected by the characteristics of the surface wave field, the basinscale background currents, the particle properties of the upper sediment layer and the water
level. Nearshore hydrodynamics as well as sediment dynamics and budget were studied by
short- and long-term measurements in alpine Lake Constance. These measurements were
combined with different numerical experiments using the wave model SWAN, the nearshore
model SWASH and DELFT3D. Most of the time wind waves were characterized by small
amplitudes, high frequencies and short wave lengths. Only during strong onshore wind events
significant remobilization and transport of particles could be observed. The background
current field was dominated by alongshore-directed velocities especially in the shallow
nearshore zone, where significant net sediment transport occurred. The pattern of the resultant
background current field is seasonally varying and is highly affected by the interplay between
the direction of the large-scale currents, the direction of the surface wave field, and
morphological features in the nearshore zone. The interactions between shore morphology,
sediment remobilization, and the background currents are decisive for the horizontal pattern
of the net sediment budget.
KEYWORDS
lake hydrodynamics, surface waves, currents, sediment dynamics, numerical modelling
Physics of boreal lakes –reflections on our learning during the last decades
Timo Huttula
Finnish Environment Institute
Freshwater Centre
timo.huttula@ymparisto.fi
Abstract
I will summarize the experience and lessons in physical limnology in the boreal zone since 1970’s using
case studies from Finland, Estonia and Russia and even East Africa as examples. During this period the
legislation and administrative frame work in Finland was changed several times. Also global technical
development was very intensive as computational resources have grown and some new algorithms have
been developed. New innovations also in sensor development have facilitated abundant data collection.
The deepening collaboration with hydrochemists, ecologists and physicists in our country has been
intensified by administrative solutions as well as by international contacts and projects. Our
understanding has grown very much. Chemical pollution and global change are major challenges
together with eutrophication. Sensible use of human and material resources are very much needed.
Kimberly Huynh
20th International Physical Processes in Natural Waters (PPNW) Workshop
February 15, 2017
Thermally-driven transport of dissolved methane and carbon dioxide through the water
column in a subtropical rice field
Wetlands are the single largest source of methane emissions, but the underlying processes
behind this flux are not yet fully understood. Typically, methane fluxes from wetlands have been
attributed to ebullition (bubbling) and to transport through vegetation. However, a third major
pathway--hydrodynamic transport--has been seen in a temperate wetland in the
Sacramento-San Joaquin Delta. We wish to explore whether this additional pathway is also
important to a subtropical rice paddy site where the diel thermal cycle is less pronounced than in
the temperate site.
Measurements in the surface water of a rice field were collected over two different rice-growing
seasons in 2016 and 2017. Specific measurements collected included dissolved and
atmospheric methane concentration, surface water velocity, and air and water temperature.
These were used to augment a long-term dataset of micrometeorology and gas fluxes.
Together, these data demonstrate the role that surface water motions play in the fluxes between
soil and atmosphere. Data are analyzed to reveal the fraction of total methane flux that is
governed by advective/diffusive transport through surface water, and daily cycles in this
behavior. Results will be used to advance predictions of atmospheric methane gas
concentrations and could be foundational for developing methane management solutions.
Closing this gap in knowledge is key to improving calculations of current global greenhouse gas
emissions.
Hydrological controls on spring carbon gas emissions from sub-arctic lakes
Joachim Jansen1, Brett Thornton1, Martin Wik1, and Patrick Crill1
1
Department of Geological Sciences and Bolin Centre for Climate Research, Stockholm University,
Stockholm, Sweden
Northern lakes are an important atmospheric source of climate forcing trace gases — methane (CH4)
and carbon dioxide (CO2) — despite being ice covered for up to seven months of the year. As much as
52% of annual C emissions occur during ice-out in spring. This flux is driven primarily by accumulation
of carbon gas in the water column below the ice. Hydrodynamic processes such as under-ice circulation
and snowmelt events are important regulators of lake biogeochemistry. In this study we present a
detailed carbon gas budget for three subarctic lakes in northern Sweden during winter and spring. We
combine continuous eddy covariance and water temperature measurements with monthly
observations of the dissolved gas content. CH4 and CO2 accumulated at the sediment-water interface
following ice-on and at intermediate depths after anoxia set in. This accumulation pattern made the
pelagic zone the locus of the spring CH4 efflux, contrary to summer, when the largest emissions were
from the littoral zone. However, the vertical distribution of carbon gas remains somewhat puzzling.
Stable water column stratification prohibited vertical transport, with dissolved CH4, CO2 and HCO3−
instead strengthening the thermal density gradient. The process of diffusion was too slow to generate
the concentration profiles observed. We therefore hypothesize that dense, carbon-rich water moved
laterally from the littoral to the pelagic zone, vertically displacing less dense water to the ice-water
interface. Snowmelt may also have displaced under-ice water. In two of the study lakes we found that
between 65 and 84% of carbon gas had disappeared several weeks prior to ice-out during periods of
heavy snowmelt. Thus, only a small part of the potential ice-out flux was picked up by the eddy
covariance system. These observations indicate that catchment hydrology plays an important part in
regulating both winter accumulation and the spring efflux of carbon gas in seasonally ice-covered lakes.
Lake classification revisited: scaling of lake seasonal stratification.
By
Georgiy Kirillin and Tom Shatwell
Leibniz-Institute of Freshwater Ecology and Inland Fisheries, Berlin, Germany
Hutchinson and Löffler's (1956) classification of lakes based on the seasonal thermal mixing
regime has become a cornerstone of any analysis of lakes as elements of the earth surface.
Until now however the lake classification has lacked a physically sound quantitative criterion
distinguishing between two fundamental lake types: thermally stratified during a large portion
of the year (mono- and dimictic) and predominantly mixed to the bottom (polymictic). Using
the mechanistic balance between potential and kinetic energy we derived a generalized
scaling for seasonal stratification in a closed lake basin. The scaling parameter is the critical
mean basin depth, Hcrit, that delineates lakes that mix regularly from those that stratify
seasonally based on lake water transparency, lake length, and an annual mean estimate for
the Monin-Obukhov length. The scaling criterion consistently describes the mixing regime
significantly better than either the conventional unbounded basin scaling or a simple depth
threshold. Thus, the generalized scaling is universal for freshwater lakes and allows the
seasonal mixing regime to be estimated without numerically solving the heat transport
equations.
Melting of lake ice: measurements and modelling
Matti Leppäranta1 and Georgiy Kirillin2
1
Department of Physics, University of Helsinki, Helsinki, Finland;
Leibniz-Institute of Freshwater Ecology and Inland Fisheries,
Berlin, Germany
matti.lepparanta@helsinki.fi, kirillin@igb-berlin.de
2
Melting of lake ice is a complex process driven primarily by solar radiation. Ice melts at the
boundaries and in the interior, with fractions depending on the solar radiation balance and
surface heat balance. Complications arise from large variability of optical properties of ice in
the melting season. To gain more understanding of the melting process, field experiments
have been performed in Finnish lakes from the boreal zone to Arctic tundra. In particular, an
extensive research program has been carried through in Lake Kilpisjärvi in the tundra zone in
2013–2014. The surface area of the lake is 37.1 km2, and the maximum depth is 57 m. The
heat budget in the melting season was dominated by the radiation balance, and turbulent heat
fluxes were small except that occasionally sensible heat flux was large. The strong solar
radiation leads to internal melting, and under the ice water warms up resulting in convective
mixing. The radiation transfer through the ice was measured using photo-synthetically active
radiation (PAR). The data obtained will be also used as the reference of mathematical model
development.
Methane bubble growth and transport in aquatic sediments observed by micro-scale X-ray
computed tomography
Liu, University of Koblenz-Landau
Biogenic methane gas bubble formation and migration in surface aquatic sediments is an important
process for global biogeochemistry cycling at sediment-water interface. However, the mechanisms of
bubble migration in sediment are still unclear. A long-term (20 d) laboratory incubation was done to
study methane bubble growth and migration mechanisms in homogenized natural sediments (clay,
sand). During the incubation experiment, X-ray computed microtomography (micro-CT) was employed to
track bubble formation dynamics. At the end of bubble growth experiment, two micro-CT column scans
were done to track bubble migration patterns in sediment in response to a scheduled water level
change. The incubation shows capillary invasion and sediment expansion were both important in bubble
growth in the two investigated sediments. Associated with sediment expansion, a significant gasenriched upper layer (8 cm) was observed in sand. Bubbles were observed to move only in the surface
layer of sand, in contrast to the entire depth in clayey sediment. Bubble migration in sediments was
primarily determined by the mobility of bubbles, which was determined by the relative size of pores (in
sediment) and bubbles. The findings will provide a solid basis for a methane bubble release model in
sediments.
20th Workshop on Physical Processes in Natural Waters, Helsinki, Finland, 21-25 August 2017
Near-surface turbulence and gas exchange velocities in shallow
streams
A. Lorke1*, C.Noss1, P. Bodmer1, K. Koca1
1
Institute for Environmental Sciences, University of Koblenz-Landau, Landau, Germany
*Corresponding author, e-mail lorke@uni-landau.de
ABSTRACT
We analyzed the relationships between surface flow type, near-surface flow and turbulence,
and the air-water gas exchange velocity (k600) in shallow streams. The surface flow type is an
empirical measure of the spatial variation of hydraulic conditions in streams, which is often
mapped visually for ecological and morphological surveys. We used freely floating
instrumented particles, which were equipped with acceleration sensors to obtain a
measurement-based categorization of surface flow types. The spatial and temporal dynamics
of near surface flow fields and turbulence were measured using acoustic Doppler velocimeter
and particle image velocimetry. Gas exchange velocities were measured for carbon dioxide
and methane with flux chambers. All measurements were conducted in natural streams and
cover a range of different surface flow types. The data analysis focused on improving the
current understanding of the processes and driving forces of local turbulence generation and
gas exchange velocities in shallow streams. We further explored the potential of using surface
flow type characterization and measurement for estimating gas exchange velocities at the
reach scale and beyond.
Circulation and Respiration in Ice-covered Alaskan Arctic Lakes
Sally MacIntyre1,2*, Alicia Cortes Cortes2, and Steven Sadro 2
1
Department of Ecology, Evolution and Marine Biology, University of California at Santa
Barbara, Santa Barbara, CA
2
Marine Science Institute, University of California at Santa Barbara, Santa Barbara, CA
Abstract - Arctic lakes are ice-covered 9 months of the year. For some of this time, the sediments
heat the overlying water, and respiration in the sediments increases specific conductivity,
depletes oxygen, and produces greenhouse gases (GHG). Whether anoxia forms and whether the
greenhouse gases are sequestered at depth depends on processes inducing circulation and upward
fluxes. Similarly, whether the GHG are released at ice off depends on the extent of vertical
mixing at that time. Using time series meteorological data and biogeochemical arrays with
temperature, specific conductivity, and optical oxygen sensors in 5 lakes ranging from 1 to 150
ha, we illustrate the connections between meteorological forcing and within lake processes
including gravity currents resulting from increased density just above the sediment water
interface and internal waves including those induced by winds acting on the surface of the ice
and at ice off. CO2 production was well predicted by the initial rate of oxygen drawdown near
the bottom at ice on. Upward density flux depended on lake size, with values initially high in all
lakes but near molecular in lakes of a few hectares in size by mid-winter. Both CO2 production
and within lake vertical fluxes were independent of the rate of cooling in fall and subsequent
within lake temperatures under the ice. Anoxia formed near the sediments in all 5 lakes with the
concentration of CH4 dependent, in part, on lake size and depth. Considerable internal wave
induced mixing occurred at the time of ice off. Twenty to fifty percent of the greenhouse gases
produced under the ice remained in the lakes by the time thermal stratification was established in
summer despite the mixing, unless, in the case of the largest of the lakes studied, strong winds
occurred while the lake was partially ice covered. These observations and analysis lay a
framework for understanding the links between within lake hydrodynamics, within year
variability, and the fraction of greenhouse gases produced over the winter which evade at ice off.
Carbon dioxide fluxes in tropical waters: application of a surface renewal model based on
near surface turbulence and vertical mixing
Melack, University of California Santa Barbara
Exchange of carbon dioxide between surficial water and overlying atmosphere depends on the
concentration gradient and on physical processes at the interface, usually parameterized as a gas
exchange coefficient (denoted as k). Near-surface concentrations of the gases depend on
concentrations at different depths and physical processes that bring them to the air-water
interface. We report results from studies in an Amazonian lake and reservoir in which we used a
combination of meteorological data, time series measurements of temperature and oxygen with
high resolution moored sensors, carbon dioxide concentrations and chamber-based gas fluxes,
and a temperature-gradient microstructure profiler used to compute the rate of dissipation of
turbulent kinetic energy through the water column. Surface energy budgets and near-surface
dissipation rates derived from the microstructure profiler are used to calculate k values with a
surface renewal model; k values are also obtained by inverse procedures from chamber
measurements. In contrast to expectations, our results indicate that k values can be high under
light winds and heating. For winds less than 3 m s -1, there was no dependence on wind speed
under heating, and gas transfer coefficients were about 4 times higher than predicted from
established relations with wind speed. By combining these results with measurements of gas
concentrations and meteorological variables, we estimate that fluxes of carbon dioxide from
Amazon floodplains are likely to be 2 to 3 times greater than previous estimates.
Riverine carbon and nitrogen & Greenhouse gases (GHGs) emissions in rivers of the Tibetan
Plateau
Qu, Lappeenranta University of Technology, Finland
The permafrost soils on the Tibetan Plateau represent a carbon store of 12.3 Pg-C (Pg=1015g),
which is potentially vulnerable to climate warming. It was revealed due to climate change, old
carbon from permafrost regions of the Tibetan Plateau is being exported with its melting and
degradation, and there will be an increasing export of ancient carbon from the rivers if the
temperature constantly increasing on the plateau. Once previously frozen permafrost carbon is
released into surface waters, it will likely be rapidly degraded and result in CO 2 evasion to the
atmosphere. The pCO2 in the rivers of the Tibetan Plateau were significant related to DIC. And
during the summer half year, the fluxes of CO 2 (3,452 mg-C m2 d-1) are comparable with most other
rivers in the world, despite the low partial pressures of CO 2 in rivers of the plateau. Therefore, with
global warming, an increasing exportation of carbon substances (including CO 2 emissions) from
rivers of the Tibetan Plateau can be expected in the future, which will potentially add feedback to
the regional climate.
Lake CO2 measurements using UAV
Sahlée, Uppsala University
Abstract
Recent technological advancements of miniaturized sensors, electronics and the development
of navigational software now allows for meteorological measurements to be performed on the
local (i.e. lake) scale using small unmanned aerial vehicles (UAVs). These platforms are
relatively cheap, easy to deploy and maintained in comparison to manned aircraft, and has the
potential to revolutionize studies of atmospheric boundary layer processes. Here we present
atmospheric CO2 measurements from a Swedish lake using a recent prototype drone system
consisting of a commercially available quadcopter and an electronics hub produced by the
Swedish company Sparv Embedded.
PPNW 2017 Workshop Abstract
Verlet-Banide Antonin
Uppsala University
Methane outgassing observation from lake Erken
Uppsala University is involve in a interdisciplinary project focusing on the impact of inland
water ecosystems in the global carbon cycle. As part of the meteorology group my aim is to
focus on methane outgassing from lake Erken via a field campaign combining Eddy covariant
measurements and water methane concentration measurements.
Lake Erken is located in east central Sweden, about 70 km east of Uppsala (59o 51’ 00" N, 18o
34’ 00" E). The water lake surface is 24 km2 with a mean depth of 9 m
The aim of this field campaign is to asses the variation in methane concentration in the water
other a week period. The water measurement will be done continuously overs 12 h observation
period. Simultaneously the methane flux outgassing from the lake will be observe using an Eddy
covariance method on an island locate close to the southeast of the lake. The measurements
are taken on a two level tower. First level at 4.10 m and the second level at 6.17 m height both
with measurements of wind speed, wind direction and temperature. CH4 flux measurement are
at the first level with a sonic anemometer for wind measurements and a Li-7700 (LI-COR
Inc.,Lincoln,NE,USA) for the methane concentration.
From the result a long-term continuous methane flux observation is attempted. In addition, the
methane water concentration variation will be study. I will try to observe the potential
concentration variation due to different meteorological process. I will also try to observe if
there is a diurnal variation in water methane concentration.
From the the results of the methane concentration and the flux measurement a first attempt to
estimate the transfer velocity will de calculated from those continuous measurements. I will try
to understand the the impact of different physical process on the estimated transfer velocity.
Contribution of high and low frequency internal waves to boundary turbulence in a
lake
Wain, University of Bath
Stratification in lakes restricts vertical mixing and often controls the spatial variability of
nutrients and other substances, affecting the distribution of dissolved oxygen in the water
column, the availability of nutrients to phytoplankton, and transport of pollutants between the
hypolimnion and epilimnion. The interior of lakes is often quiescent and most of the mixing
in a lake occurs at the sloping boundaries, where wind-induced internal waves create
turbulence (which leads to mixing) through interactions with the lakebed. To predict the
occurrence and strength of turbulence in terms of meteorological forcing and stratification,
we investigated the dependence of internal wave type, and their contribution to turbulence on
the slope, on the Lake number, which compares the stabilizing tendency of stratification to
the destabilizing tendency of the wind.
Three thermistor chains and a meteorological station were deployed in West Okoboji Lake
(length ~ 9 km, max. depth ~ 40 m) for two weeks. A wavelet analysis was conducted to
determine time periods when different wave frequencies were excited, with particular focus
on the first vertical mode seiche, the critical frequency with respect to the stratification and
slope, and high frequency waves in the band of 1-10 times the buoyancy frequency. We
measured the velocities in the bottom boundary layer (BBL) with a high resolution acoustic
current profiler (2 MHz Nortek HR Aquadopp) and then computed the turbulent dissipation
rate using the structure function method, which uses the spatial correlations of velocity along
a beam to estimate the dissipation. This generated a two week time series of turbulent
dissipation rate in the BBL which was then compared to the wavelet amplitudes.
During the deployment, a strong daily wind forced near constant internal wave activity. The
theoretical period of the first vertical mode seiche was ~17 hours, but the diurnal wind
forcing interfered with free oscillation of this mode. Although not an obvious natural
frequency of the lake, waves of the critical frequency (which had a period of ~11 hours) were
activated throughout the measurement period. High-frequency waves were observed in the
thermistor chain near the slope at the lowest Lake number wind events. The turbulence
observed on the boundary was highest during these events, implying that the low frequency
seiching was less important than higher frequency motions in driving turbulence on the slope.
High spatial variability in stream gas transfer velocity revealed by ADV
derived turbulence measurements
Marcus Wallin1, Jovana Kokic2, Erik Sahlée1, Dominic Vachon3, and Sebastian Sobek2
1
Dept. of Earth Sciences, Program for Air, Water and Landscape Sciences, Uppsala University, Sweden
2
Dept. of Ecology and Genetics/Limnology, Uppsala University, Sweden
3
Dept. F.-A. Forel for Environmental and Aquatic Sciences, University of Geneva, Switzerland
Abstract
Streams are concluded to be major emitters of carbon dioxide (CO2) and methane (CH4)
to the atmosphere. However, streams are also known to be very variable in their emission
rates resulting in large uncertainties connected to scaled emission estimates. One of the largest
uncertainties derives from the choice of an appropriate gas transfer velocity (k), which
describes the physical efficiency for gas exchange across the stream-atmosphere interface. In
this study we therefore aimed to determine the variability in k and subsequent emission rates
within and across streams of different stream order (SO) using direct turbulence
measurements by an Acoustic Doppler Velocimeter (ADV). The measured ADV-dissipation
rates were converted into gas transfer velocities which were used to calculate CO2 and CH4
emissions. The results show that ADV measurements can be used to determine k in streams.
There was high spatial variability in k and corresponding emissions at small scales, both
within stream reaches and across SO, and especially during high discharge. There was further
no clear relationship between k and SO nor specific stream characteristics such as width and
depth, which are parameters often used in empirical models of k. This suggests that the
relationship between physical stream characteristics and k are not straightforward, and may
introduce uncertainties when used for scaling purposes. Improved understanding of the smallscale variability in the physical properties along streams, especially during periods of high
discharge, is therefore an important step to reduce the uncertainty in existing gas transfer
models and GHG emissions estimates for stream systems.
20th Workshop on Physical Processes in Natural Waters, Hyytiälä Forestry Station, Finland, 21-25 August 2017
Bacteria induced mixing – comparing field observations with DNS
A. Wüest1,2, G. Constantinescu3, O.R. Sepulveda Steiner2, T. Tokyay4 and T. Sommer1,5
1
2
3
Physics of Aquatic Systems Laboratory - Margaretha Kamprad Chair of Environmental Science
and Limnology, ENAC, EPFL, Lausanne, Switzerland
Department of Civil and Environmental Engineering, The University of Iowa, Iowa City, USA
4
5
Eawag, Surface Waters - Research and Management, Kastanienbaum, Switzerland
Department of Civil Engineering, Middle East Technical University, Ankara, Turkey
Center for Integrated Building Technology, Lucerne School of Engineering and Architecture, Horw, Switzerland
ABSTRACT
In this presentation we extend earlier reports on this workshop about the ability of purple
sulphur bacteria Chromatium okenii (C. okenii) to convectively mix their natural stratified
environment over vertical scales of several dm to m. For this type of bioconvection, the
microorganisms induce water movements not by their own propulsion as such, but by their
ability to swim upwards and thereby to change the vertical density structure of the waterbacteria mixture. As the C. okenii are denser than water (~1.15 g mL-1), their drive to swim
towards the light generates a homogenised water layer of macroscopic dimensions. In this
particular study in a meromictic alpine lake, a homogeneous bacterial layer develops each
summer in ~12 m depth and its vertical thickness ranges from 0.3 to 1 m.
Using Direct Numerical Simulation (DNS), we have been able to model the expansion
dynamics of this homogenised layer for realistic environmental conditions, characterized by a
narrow transition between oxic (upper half) and anoxic sulfidic (lower half) lake water, and C.
okenii densities of ~104 to ~105 cells per mL and upward swimming speeds of ~9 μm s-1. DNS
results show that the buoyancy flux generated by the upward swimming bacteria balances well
with the change of the potential energy related to the expansion of the mixed layer and the
dissipation of turbulent kinetic energy. The dissipation of the convective eddies, measured by
using a RSI VMP-500 microstructure profiler operated from a float, showed excellent
agreement with DNS predictions. This consistency between field observations and modelling,
both from a conceptual as well as from a numerical point of view, strongly suggests that this
outstanding, and quite unique, phenomenon in natural waters is adequately interpreted.
LIST OF PARTICIPANTS
First name
Joonatan
Marina
Mahan
Lais
Bertram
Sergei
Damien
Alicia
Daphne
Rebecca
Sabine
Andrew
Alexander
Daniela
Andres
Sofya
Hilmar
Jussi
Timo
Kimberly
Hiroki
Joachim
Helena
Ji-Hyeon
Georgiy
Petri
Charles
Matti
Madis-Jaak
Liu
Gregorio Alejandro
Andreas
Jun
Sally
Ivan
Roseanne
Daniel
John
Anne
Andrés Felipe
Maria
Bin
Berit
Miitta
Ricardo
Vera
Erik
Stefano
Uwe
Victor
Dominic
Antonin
Timo
Sergei
Danielle
Marcus
Michael
Alfred Johny
Zheng-Jian
Last name
Ala-könni
Amadori
Amani Geshnigani
Amorim
Boehrer
Bogdanov
Bouffard
Cortes
Donis
Ellis
Flury
Folkard
Forrest
Franz
Gomez Giraldo
Guseva
Hofmann
Huotari
Huttula
Huynh
Iwata
Jansen
Jäntti
Kim
Kirillin
Kiuru
Lemckert
Leppäranta
Lilover
Liu
López Moreira Mazacotte
Lorke
Ma
MacIntyre
Mammarella
McDonald
McGinnis
Melack
Ojala
Posada Bedoya
Provenzale
Qu
Rabe
Rantakari
Roman-Botero
Rostovtseva
Sahlée
Simoncelli
Spank
Stepanenko
Vachon
Verlet-Banide
Vesala
Volkov
Wain
Wallin
Weber
Wüest
Yang
Affiliation
University of Helsinki, Finland
University of Trento, Italy
University of Bath, UK
University of São Paulo, Brazil
Helmholtz Centre for Environmental Research (UFZ), Germany
Northern Water Problems Institute, Russian Academy of Sciences, Russia
Eawag, Switzerland
University of California Santa Barbara (UCSB), USA
University of Geneva, Switzerland
University of Bath/Plymouth Marine Laboratory, UK
Ecole Polytechnique Fédérale de Lausanne (EPFL), Switzerland
Lancaster University, UK
University of California Davis, USA
University of Helsinki, Finland
Universidad Nacional de Colombia, Colombia
Lomonosov Moscow State University, Russia
University of Konstanz, Germany
University of Helsinki, Finland
Finnish Environment Institute (SYKE), Finland
University of California Berkeley, USA
Shinshu University, Japan
Stockholm University, Sweden
University of Eastern Finland, Finland
Université du Québec à Montréal (UQAM), Canada
Leibniz-Institute of Freshwater Ecology and Inland Fisheries, Germany
Finnish Environment Institute (SYKE), Finland
University of Canberra, Australia
University of Helsinki, Finland
Tallinn University of Technology, Estonia
University of Koblenz-Landau, Germany
University of Trento, Italy/Leibniz-Inst. of Freshwater Ec. and Inland Fisheries, Germany
University of Koblenz-Landau, Germany
Hubei University of Technology, China
University of California Santa Barbara (UCSB), USA
University of Helsinki, Finland
Centre for Ecology & Hydrology, UK
University of Geneva, Switzerland
University of California Santa Barbara (UCSB), USA
University of Helsinki, Finland
Universidad Nacional de Colombia, Colombia
University of Helsinki, Finland
Lappeenranta University of Technology, Finland
Marine Laboratory, Marine Scotland Science, UK
University of Helsinki, Finland
Universidad Nacional de Colombia, Colombia
P. P. Shirshov Institute of Oceanology, Russian Academy of Sciences, Russia
Uppsala University, Sweden
University of Bath, UK
TU-Dresden, Germany
Lomonosov Moscow State University, Russia
University of Geneva, Switzerland
Uppsala University, Sweden
University of Helsinki, Finland
Northern Water Problems Institute, Russian Academy of Sciences, Russia
University of Bath, UK
Uppsala University, Sweden
Helmholtz Centre for Environmental Research (UFZ), Germany
Eawag and EPFL, Switzerland
Hubei University of Technology, China
e-mail
joonatan.ala-konni@helsinki.fi
marina.amadori@unitn.it
mag63@bath.ac.uk
laisamorim@usp.br
bertram.boehrer@ufz.de
sergey.r.bogdanov@mail.ru
damien.bouffard@eawag.ch
alicia.cortes@ucsb.edu
daphne.donis@unige.ch
re346@bath.ac.uk
sabine.flurymcginnis@epfl.ch
a.folkard@lancaster.ac.uk
alforrest@ucdavis.edu
daniela_franz@gmx.de
eagomezgi@unal.edu.co
guseva.sofya.pavlovna@gmail.com
hilmar.hofmann@uni-konstanz.de
jussi.huotari@helsinki.fi
timo.huttula@ymparisto.fi
kim.huynh@berkeley.edu
hiwata@shinshu-u.ac.jp
joachim.jansen@geo.su.se
helena.jantti@uef.fi
jihyeonkim91@gmail.com
kirillin@igb-berlin.de
petri.kiuru@ymparisto.fi
charles.lemckert@canberra.edu.au
matti.lepparanta@helsinki.fi
madis-jaak.lilover@msi.ttu.ee
liu@uni-landau.de
ga.lopez@unitn.it
lorke@uni-landau.de
majun150@hotmail.com
sally@eri.ucsb.edu
ivan.mammarella@helsinki.fi
rosdon23@ceh.ac.uk
daniel.mcginnis@unige.ch
melack@bren.ucsb.edu
anne.ojala@helsinki.fi
afposadab@unal.edu.co
maria.provenzale@helsinki.fi
Bin.Qu@student.lut.fi
b.rabe@marlab.ac.uk
miitta.rantakari@helsinki.fi
ricardo.rb.academico@gmail.com
vera@ocean.ru
erik.sahlee@met.uu.se
s.simoncelli@bath.ac.uk
uwe.spank@tu-dresden.de
vstepanenkomeister@gmail.com
dominic.vachon@unige.ch
antonin.verlet-banide@geo.uu.se
timo.vesala@helsinki.fi
taranarmo@gmail.com
d.j.wain@bath.ac.uk
marcus.wallin@geo.uu.se
michael.weber@ufz.de
alfred.wueest@eawag.ch
656637841@qq.com