Thais Palumbo 2023

International Soil and Water Conservation Research xxx (xxxx) xxx
Contents lists available at ScienceDirect
International Soil and Water Conservation Research

journal homepage: www.elsevier.com/locate/iswcr
Original Research Article
Best management practices to reduce soil erosion and change water

balance components in watersheds under grain and dairy production

Thais Palumbo Silva a, Danielle Bressiani b, Ederson Miguel Reichert c, *
Diniz Ebling a, Jose
a
Soil Science of Federal University of Santa Maria, Roraima Avenue, number 1000, CEP: 97.105-900, Santa Maria, Rio Grande do Sul, Brazil
b
Technological Development Center, Federal University of Pelotas, Gomes Carneiro Street, number 1, CEP: 96010-610, Pelotas, Rio Grande do Sul, Brazil
c rio s/n, CEP 97.105-900, Santa Maria, Rio Grande do Sul, Brazil
Department of Soil, Federal University of Santa Maria, Campus Universita
a r t i c l e i n f o a b s t r a c t
Article history: Soil erosion and sedimentation are among the most serious global environmental problems. Soil and
Received 4 July 2022 water conservation measures have been proven effective ways to reduce soil loss. The objective of this
Received in revised form study was to evaluate the impact of three approaches of soil and water conservation measures (soil
20 May 2023
management, vegetative measures, and structural practices) on soil erosion and water balance of two
Accepted 9 June 2023
Available online xxx
paired agricultural watersheds located in Southern Brazil. Streamflow and sediment monitoring was
carried out from 2016 to 2019 in the two small paired agricultural watersheds; called North watershed
(NW) and South watershed (SW). Modeling using Soil & Water Assessment Tool (SWAT) was performed
Keywords:
Agricultural hydrology
to simulate individual (nine scenarios) and combined (four scenarios) best management practices
Conservation measures (BMPs), by including the three approaches. Among the nine individual BMP scenarios, the most effective
SWAT in reducing soil erosion was crop rotation and cover crop (sediment yield, SY, reduction of 38.4 for NW,
Subtropical watersheds and 28.8% for SW). Among the four combined scenarios, the association of all conservation approaches
was the most effective in reducing soil erosion (SY reduction of 46 for NW, and 41.5% for SW), followed by
the vegetative measures scenario (SY reduction of 43.5 and 34.1% for NW and SW). All combined sce-
narios increased infiltration and subsurface water components, and decreased surface runoff. The
findings of this study can help farmers and policymakers choosing appropriate BMPs to reduce current
soil erosion problems and promote water and soil conservation.
© 2023 International Research and Training Center on Erosion and Sedimentation, China Water and
Power Press, and China Institute of Water Resources and Hydropower Research. Publishing services by
Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article under the CC BY-
NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction South America, Southeast Asia, and Sub-Saharan Africa, where

there is intense agriculture (Borrelli et al., 2017). Compared to
Soil erosion is one of the most serious global environmental countries in South America, the effects of soil erosion are severe in
problems. According to the Food and Agriculture Organization (FAO Brazil. Besides intense agriculture, Brazil has excessive soil erosion
Food and Agricultural Organization of the United Nations, 2019), rates due to the rainfall regime, soil properties, and slope charac-
accelerated soil erosion by overgrazing, intensive agriculture, and teristics (Guerra et al., 2014), but the country is also considered a
deforestation can increase soil loss. During the mid-1990s, about pioneer in the adoption of soil conservation practices (Landers,
30% of the world’s cultivated land has become unproductive 2005). Soil erosion and water balance can be impacted by crop-
(Pimentel, 2006). However, if nothing is done to minimize soil land activities, and soil and water conservation measures are key to
erosion, over 90% of the world’s cultivated land could become improve the negative changes (Boufala et al., 2021; Li et al., 2021;
degraded in 2050 (FAO Food and Agricultural Organization of the Reichert et al., 2017; Strauch et al., 2013; Uniyal et al., 2020).
United Nations, 2019). Soil loss rate has been increasing mainly in In the late 1970s, the FAO of the United Nations and the Brazilian
Agricultural Research Corporation (EMBRAPA) started a soil con-
servation project in two representative watersheds from Plateau
* Corresponding author. region of South Brazil. This region has a particular characteristic
E-mail addresses: daniebressiani@gmail.com (D. Bressiani), ederdinize@gmail. related to highly-weathered soils, high rainfall erosivity (Oliveira
com
(E.D. Ebling), thaispalumbosilva@hotmail.com, reichert@ufsm.br
(J.M. Reichert).
et al., 2013), and intense agriculture, which contribute to severe
https://doi.org/10.1016/j.iswcr.2023.06.003
2095-6339/© 2023 International Research and Training Center on Erosion and Sedimentation, China Water and Power Press, and China Institute of Water Resources and Hydropower Research.
Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Ebling et al., Best management practices to reduce soil erosion and change water balance
Please cite this article as: T.P. Silva, D. Bressiani, E.D.
components in watersheds under grain and dairy production, International Soil and Water Conservation Research, https://doi.org/10.1016/
j.iswcr.2023.06.003
Ebling et al.
T.P. Silva, D. Bressiani, E.D. International Soil and Water Conservation Research xxx (xxxx) xxx
soil erosion. The focus of the project was to promote conservation an agricultural watershed under no-till, and the most effective
agriculture to minimize the sedimentation in reservoir Passo Real. scenario to reduce soil erosion included all the adopted approaches
The Passo Real Reservoir was built in the 1970s, which is the largest of conservation measures (crop rotation, contouring farming,
artificial lake in the Rio Grande do Sul State, and it is important for terracing, and riparian forest). Therefore, few studies have evalu-
energy generation and public water supply (Peixoto et al., 2017). ated the impact of the three conservation measures approaches
However, the project was abandoned and, four decades later, the separately and not only on soil erosion, but also on the water bal-
monitoring restarted associated with modeling. Modeling has been ance components.
widely used in soil erosion studies because it is an alternative tool In South Brazil, farmers have been implementing no-till alone
that provides estimates of the effect of best management practices rather than conservation agriculture (Reicosky, 2015). However,
(BMPs) before their implementation at a watershed scale (Briak severe soil erosion and sedimentation problems have been occur-
et al., 2019; Didone et al., 2017; Strauch et al., 2013; Uniyal et al., ring in the plateau region of South Brazil, mainly on upstream
2020). watersheds to Passo Real reservoir (Broetto et al., 2017; Ebling,
Studies focused on soil and water conservation measures have 2018). To minimize soil erosion and its negative consequences,
been carried out across the world (Afroz et al., 2021; Berihun et al., the association of conservation measures, between soil, vegetative
2020; Briak et al., 2019; Didone et al., 2017; Gashaw et al., 2021; and structural BMPs may improve soil properties, topography and
Ricci et al., 2020; Strauch et al., 2013). The knowledge of the effect structure, decrease surface runoff and sediment yield, and thus
of soil and water conservation practices provides essential infor- reduce reservoir sedimentation. In general, understanding the
mation for adequate management of land use. Conservation prac- impacts of different conservation measures in these representative
tices are broadly divided into soil management, vegetative agricultural watersheds is key to help farmers and decision-makers
measures and structural practices (Bertoni & Lombardi Neto, 2014). in choosing feasible and appropriate BMPs to reduce soil erosion
These practices can help minimize soil erosion, in which the soil problems, both on- and off-site.
management practices increase soil infiltration rate by improving We evaluated the effectiveness of different conservation mea-
soil structure, the vegetative measures decrease the impact of sures approaches (soil management, vegetative measures, and
raindrops by covering the soil surface, and the structural tech- structural practices) in two agricultural paired watersheds. We
niques reduce the velocity and volume of surface runoff by modi- hypothesized that the association of all BMPs approaches is the
fying the landscape topography. Uniyal et al. (2020) evaluated the most effective to reduce soil erosion. This study was carried out
effect of vegetative and structural practices in an Indian watershed with the following objectives: to quantify streamflow and sediment
and showed that structural BMPs were more effective in reducing yield in agricultural paired watersheds and identify critical sub-
sediment yields and surface runoff than vegetative BMPs. basins under current management by using the SWAT model; to
Conversely, Laufer et al. (2016) implemented only vegetative assess the effectiveness of individual and combined BMPs for
measures on farm trials in Southern Germany, and could reduce controlling soil erosion, based on the different BMPs approaches;
98% of soil loss compared to base conditions of intense tillage. and to simulate the effects of combined BMPs on water balance
Himanshu et al. (2019) implemented different soil management components.
scenarios in an Indian watershed using the SWAT model, such as
conservation tillage, zero tillage, and field cultivation, and observed
2. Material and methods
a reduction of 9% in sediment yield with respect to the conventional
tillage simulation.
2.1. Study area
Several studies have demonstrated that cropland areas can
impact the water balance components (Reichert et al., 2017), such
The study was conducted in two paired watersheds located in
as infiltration (Sun et al., 2018), groundwater recharge (Yifru et al.,
the physiographic plateau region in southern Brazil (state of Rio
2021), evapotranspiration (Hu et al., 2021), soil water content
Grande do Sul, RS). The watersheds drain directly into the reservoir
(Lopes et al., 2021; Mallet et al., 2020), and water yield (Hu et al.,
Passo Real, the largest of RS, with over 225 km2 in surface area
2021; Lopes et al., 2021). Then, the conservation measures posi-
(Fig. 1). This reservoir comprises a system of energy generation with
tively impact the water balance components (Boufala et al., 2021;
a power of 158 MW. These paired watersheds were chosen for
Freitas et al., 2021; Li et al., 2021; Uniyal et al., 2020). Uniyal et al.
being representative, because both have tropical well-weathered
(2020) showed that different conservation measures could
soils, land uses and relief typical of southern Brazil under crop
decrease surface runoff and increase lateral flow, aquifer recharge,
and livestock production, where the studied watersheds are
baseflow, and water percolation. Similar results were found by
dominated by grain and dairy production, to characterize the
Boufala et al. (2021), in which conservation practices reduced sur-
magnitude of soil erosion and hydrological processes in similar
face runoff and increased the baseflow. Freitas et al. (2021) showed
conditions in terms of land use, soil, and climate, which differ in
that terracing increased the soil available water to plants in no-till
size and shape, percentage of cropland areas, and riparian vegeta-
system.
tion (Fig. 1). Based on spatial position, one is called the North
Although no-tillage is implemented to reduce soil erosion, the
watershed (NW) (28 45'17.73"S and 53 6'11.12" W) and the other
effectiveness of this practice alone is insufficient to control soil loss
et al., 2019; Londero et al., 2018). When no-tillage is South watershed (SW) (28 45'34.27" S and 53 6'28.83" W). The
(Didone
drainage areas of NW and SW are 0.94 km2 and 0.54 km2,
associated with a structural practice, such as terraces, these prac-
respectively.
tices have shown efficiency to decrease sediment yield (Londero
According to Ko € ppen’s classification, the climate type is Cfa, i.e.,
et al., 2018). Yet, the association of the three conservation mea-
subtropical humid without a dry season, with an average annual
sures proposed (soil management, vegetative measures, and
rainfall of 1,750 mm and an average temperature of 18 C (Alvares
structural practices) can provide better results to minimize soil
et al. et al., 2013). The geological bedrock is basaltic, with deep, highly
erosion and improve water available. For example, Didone
weathered soils (Ferrasols and Nitisols). The soils were classified
(2017) evaluated the impact of different BMPs on soil erosion in
according to the World Reference Base (WRB/FAO, 1998) as
2
Ebling et al.
Fig. 1. Location of the North (NW) and South (SW) watersheds and their maps about (a) elevation, (b) soil, (c) slope, and (d) land use. Source for elevation: ASF Data Search (2019),
and for soil survey: Tornquist (2007).
Ferrasols, Leptosols, Nitisols, Acrisols, and Eutric Gleysols (Fig. 1b). sediment manually collected samples during rainfall events using a
The landscape includes gentle slopes (<8%) and hillside slopes with USDH-48 sampler (Ebling, 2018). Finally, SY was estimated by
higher steepness (>15%) (Fig. 1c). Land use in both watersheds multiplying SF and SSC at every 10-min interval.
consists of native forest, cropland, and cultivated pasture. The main
crops in agricultural areas are soybean (Glycine max) and maize (Zea
mays), and the cultivated pasture consists of Tifton 85 (Cynodon 2.3. Modeling hydrological and soil erosion processes
dactylon) (Fig. 1d).
These paired watersheds are managed under family farming 2.3.1. Model setup
systems, which are characterized by small production areas under Hydrological and soil erosion processes were simulated using
intensive grain cultivation under no-till. The no-till system in these the Soil and Water Assessment Tool (SWAT), which is a process-
areas only keeps the ground covered by previous crops and it was based, semi-distributed, and continuous-time model (Arnold
adopted in late 1970’s with the implementation of the conservation et al., 1998). It was developed to predict the impact of manage-
project. ment practices on water, sediment, and agricultural chemical at a
watershed scale. SWAT has been largely applied in the world and in
Brazil (Bressiani et al., 2015; Gassman et al., 2014). SWAT model
2.2. Hydrological and soil erosion monitoring requires a large amount of spatial (i.e., digital elevation model
-DEM, land use, and soil maps) and temporal (meteorological pa-
The watersheds were monitored for four years, from April 2016 rameters and crop management) data to simulate different physical
to December 2019. Rainfall (R), streamflow (SF), and sediment yield processes. The DEM was used to create the slope map and discretize
(SY) were measured using automatic sensors recorded at 10-min water bodies and subbasins. In addition to DEM, Land use and Soil
intervals. Each automated measuring station consisted of a maps were used to define Hydrological Response Units - HRUs. For
spillway located in a watershed outlet (Fig. 1) with a rainfall gauge, NW and SW, the areas of 0.94 km2 and 0.54 km2 have been dis-
a water level sensor (limnigraph), and a turbidity sensor (turbi- cretized into 17 subbasins with 350 HRUs, and 10 subbasins with
dimeter) connected to a data logger. The SF was estimated from 268 HRUs, respectively.
water level measurements by the conversion of pressure values The DEM data was obtained from Alaska Satellite Facility (ASF,
into streamflow using the appropriate discharge rating curve 2019) with a spatial resolution of 12.5 m. Then, a DEM was gener-
calculated for the monitoring section. The streamflow, for each ated with a spatial resolution of 5 m by contouring vectors. Land
spillway, was calculated using equations developed by Ebling et al. use maps were created from the interpretation and classification of
(2022). The main type of sedimentation in these watersheds is of Landsat satellite images (Ladsat8/OLI images) from 2016 with
suspended sediment due to the predominance of clay soils and spatial resolution of 15 m, and confirmed in a field survey. Land use
interril erosion (Morgan, 2009). Suspended sediment concentra- information about management and parameters were added to
tion (SSC) was determined using a turbidimeter that automatically SWAT input management (.mgt). In general, the base land use
measured water turbidity by scattering of light. Turbidimeters were management was a monoculture (varying with soybean and corn)
calibrated at monthly intervals. Turbidity values were converted with a cover crop (oats) between the harvest and planting of the
into NTU (Nephelometric Turbidity Unit), and the NTU was con- main crop. Soil Map was obtained by Tornquist (2007) with a scale
verted into SSC using a calibration curve obtained by suspended of 1:10,000, and the main soil properties (soil granulometry, bulk
3
Ebling et al.
density, total porosity, available water capacity, and saturated hy- (2014), the various conservation measures can be described un-
draulic conductivity for each soil horizon) were obtained by Ebling der three main approaches:
(2018) and added to SWAT user soils databases.
The model requires continuous long-term meteorological data, - Soil Management: uses measures to prepare the soil to promote
such as precipitation, temperature (maximum and minimum), better plant growth and improve its structure; becoming the soil
wind speed, solar radiation, and relative humidity. Daily records more resistant to erosion, such as conservation tillage, residue
from 2014 to 2019 of these climate data were obtained at a station management, and organic fertilizer application (SOIL);
13 km away from the watersheds, located in the municipally of - Vegetative Measures: utilize the role of vegetation to protect the
, collected by the Brazilian National Institute of Meteorology
Ibiruba soil against the raindrop and surface runoff impacts, such as
(INMET, 2020). Then, the SWAT model calculated all hydrological crop rotation, cover crop, and strip cropping (VEG);
and sedimentological parameters. The surface runoff was simulated - Structural Practices: often involving engineering structures that
using the Soil Conservation Service (SCS) curve number (CN) improve the surface topography to control the surface runoff,
method (Soil Conservation Service, USDA 1972). Potential evapo- such as contour farming, terracing, and grassed waterways
transpiration (PET) was calculated using the Priestley-Taylor (STR).
method (Priestley & Taylor, 1972). Sediment yield was predicted
for each subbasin based on Modified Universal Soil Loss Equation In our study, nine individual BMPs were designed and tested
(MUSLE) (Williams, 1995) within the SWAT model. based on these three approaches (SOIL, VEG, and STR): residue
management (SOIL_BMP1), manure application (SOIL_BMP2),
2.3.2. Sensitivity analysis, calibration and validation conservation tillage (SOIL_BMP3), strip cropping (VEG_BMP1), crop
Daily measured streamflow and sediment yield data were used rotation and cover crop (VEG_BMP2), grazing management
for sensitivity analysis, calibration, and validation (Silva et al., (VEG_BMP3), grassed waterways (STR_BMP1), contour farming
2023). Hydrosedimentological data were measured for the period (STR_BMP2), and terracing (STR_BMP3). Summary of BMPs
from 2016 to 2019. Data from 2016, 2018, and 2019 were used for considered in this study, SWAT parameters changes, and the
calibration, and from 2017 for validation. Meteorological data of adoption criteria are provided in Table 1.
2014 and 2015 were used for model warm-up. A multi-step pro- Some studies have demonstrated that combined conservation
cedure was employed for sensitivity analysis, calibration and vali- measures are more effective in controlling erosive processes than
dation (Bressiani, 2016). The SWAT sensitivity analysis, calibration, only one adopted measure (Didone et al., 2021; Londero et al.,
and validation were performed using the Sequential Uncertainty 2018). For this reason, four BMPs scenarios were designed based
Fitting version 2 (SUFI-2) algorithm in the SWAT-CUP program on the three approaches (Soil management, Vegetative measures,
(Abbaspour et al., 2007). SUFI-2 can estimate large number of pa- and Structural practices) to assess the combined effect of BMPs on
rameters and model uncertainties in hydrological models the reduction of sediment yield (Table 2).
(Abbaspour et al., 2007). A global sensitivity analysis was per- Individual (nine conservation practices e Table 1) and combined
formed to select the most sensitivity parameters that could influ- BMPs (four scenarios e Table 2) were applied in the SWAT model to
ence the simulated outputs (streamflow and sediment yield). P- evaluate the sediment yield because there was minimal effect of
value was used to evaluate the significance of relative sensitivity, in BMPs on water yield. However, individual BMPs were evaluated at
which a p-value close to zero represents higher significance watershed scale, and combined BMPs (four Scenarios) were eval-
(Abbaspour, 2008, p. 20). Using the sensitive parameters, the cali- uated at watershed and subbasin scales for both watersheds. Lastly,
bration and validation processes were initially performed for the impacts of the combined BMPs Scenarios were simulated and
streamflow, and then for sediment yield. compared for the water balance components as surface runoff,
The performance of SWAT model was evaluated based on the aquifer recharge, percolation, evapotranspiration, and baseflow.
statistical indicators and the performance rating from Moriasi et al.
(2007) for simulation of water and sediment yield at daily time 3. Results
step. The statistical indicators were coefficient of determination
(R2), Nash-Sutcliffe Efficiency (NSE), and Percent Bias (Pbias). 3.1. Model performance
2.3.3. Individual and combined BMPs simulations Calibrated fitted values of streamflow and sediment yield pa-
SWAT model has been widely used to evaluate the effectiveness rameters are shown in Table 3. We selected fifteen and eight pa-
of the implementation of soil and water conservation measures rameters for streamflow and sediment yield calibration,
regarding watershed sediment yield in many areas of the world respectively. More results about the sensitivity analysis of param-
(Gashaw et al., 2021; Ricci et al., 2020; Strauch et al., 2013; Uniyal eters for these watersheds and calibration for different times steps
et al., 2020; Wang, Wu, et al., 2021). In this study, after calibra- for these watersheds can be found in Silva et al. (2023).
tion and validation of water and sediment yield in daily time step, The models showed satisfactory performance (based on Moriasi
several individual Best Management Practices (BMPs) and com- et al., 2007) for calibration (years of 2016, 2018 and 2019) and
bined BMPs scenarios were modeled to evaluate the effectiveness validation (2017) of streamflow and sediment yield. The observed
in decreasing soil erosion. Firstly, critical subbasins that had high and simulated average daily streamflow during the calibration and
average annual streamflow and sediment yield were identified, and validation were similar, 20.0 dm3 s1 for the North watershed and
then, individual and combined BMPs were implemented in all 25.0 dm3 s1 for the South watershed. For SY, the simulated data
subbasins to evaluate the effect of each scenario in reducing SY at was less than the observed for both watersheds. The SY averages for
watershed and subbasin scale. NW were 1.98 and 1.28 Mg yr1 for observed and simulated values
The choice of BMPs considered the four principles of soil con- during the calibration period, and 1.45 and 1.40 Mg yr1 for
servation: covering the soil to protect it from raindrop impact, observed and simulated values during the validation period,
increasing the infiltration capacity of the soil to reduce runoff, respectively. In the SW, the SY averages were 0.26 and 0.18 Mg yr1
improving soil structure, and increasing surface roughness to for observed and simulated SY values for the calibration period, and
reduce the velocity and volume of surface runoff (Bertoni & 0.49 and 0.24 Mg yr1 for the observed and simulated on the
Lombardi Neto, 2014). According to Bertoni and Lombardi Neto validation period, respectively. Model performance statistics are
4
Ebling et al.
Table 1
BMP considered, SWAT parameters changes and the adoption criteria for North and South watersheds.
Type of BMP BMP's SWAT parameters (input files) Value of BMP References Adoption criteria
Soil Management Residue CN2(.mgt) 2 Arabi et al. (2008) Soybean and Corn/All
Management OV_N(.hru) 0.2 (0.5e1 Mg ha1 of residue) soils/All slopes
(SOIL_BMP1)
Manure application FRT_KG(.mgt) 300 kg/ha Tuppad et al. (2010)
(SOIL_BMP2) FRT_SURFACE(.mgt) 0.5
Conservation tillage CH_N1 (.sub) 0.08 Tuppad et al. (2010)
(SOIL_BMP3) EFFMIX (.mgt, till.dat) 0.25
DEPTIL(.mgt, till.dat) 100 mm
CNOP(.mgt) 2
Vegetative measures Strip Cropping STRIP_N(.ops) adjusted based on the area Arabi et al. (2008)
(VEG_BMP1) STRIP_C(.ops) weighted average values for the
strips in the system
STRIP_CN(.ops) 3
STRIP_P(.ops) 0.3, for slope 0e8%; 0.35, for
slope 8e15%; 0.45, for slope
>15%
Crop rotation and Input files (.mgt) CORN/SOYBEAN/GREEN BEAN e
cover crop e
(VEG_BMP2) WHEAT eOAT e CORN/
SOYBEAN/GREEN
BEANa
Grazing GRZ_DAYS(.mgt) two 10-days cycle Vache et al. (2002) Grassland/All soils
management (except in Gleysols)/All
(VEG_BMP3) slopes
Structural Grassed waterways CH_W2(.rte) 10 Arabi et al. (2008) Soybean and Corn/Acrisols/Slope>8%
practices (STR_BMP1) CH_D(.rte) 0.6
CH_N2(.rte) 0.4
CH_COV2(.rte) 0.001
Contour Farming CONT_CN(.ops) 3 Arabi et al. (2008) Soybean and Corn/All soils/All slopes
(STR_BMP2) CONT_P(.ops) 0.5, for slope 0e8%; 0.7, for slope 8e15%; 0.9, for
slope greater than 16%
Terracing TERR_P (.ops) 0.1, for slope 2e8%; 0.14, for slope 8e15%; 0.18, for Arabi et al. (2008) Soybean and Corn/Ferralsols, Nitisols and
(STR_BMP3) slope greater than 15% ASAE (2003) Acrisols/Slope>2%
TERR_CN (.ops) 6
TERR_SL (.ops) (0.1*SLOPEþ0.9)*100/SLOPE
a- The crop rotation varied in the different years between corn, soybean, and green bean. For each crop fertilizers and pesticides were applied to improve plant growing.
Table 2
BMPs combined scenarios.
Scenarios BMP type Combinations of BMP's
Base Scenario e Without BMP's

Scenario 1 Soil Management (SOIL_BMP1)þ(SOIL_BMP2)þ(SOIL_BMP3)
Scenario 2 Agronomic measures (VEG_BMP1)þ(VEG_BMP2)þ(VEG_BMP3)
Scenario 3 Structural practices (STR_BMP1)þ(STR_BMP2)þ(STR_BMP3)
Scenario 4 Soil Management, Agronomic (SOIL_BMP1)þ(SOIL_BMP2)þ(SOIL_BMP3)þ(VEG_BMP2)þ(VEG_BMP3)þ(STR_BMP1)þ(STR_BMP2)þ(STR_BMP3)
measuresa and Structural
practices
a
Strip cropping (VEG_BMP1) was not included in Scenario 4 since its effect is similar to terracing (STR_BMP3).
summarized in Table 4. NSE and R2 of streamflow calibration and period and overestimated during low flow periods. The year of 2019
validation for both watersheds were greater than 0.5, and Pbias was a wet one for both watersheds, in which the simulated SF and
smaller than ±24.4%. For sediment yield results, the NSE and R2 SY were underestimated. For example, in the highest recorded
were greater than 0.4, and Pbias smaller than ±35.6%. According to event, on October 31, 2019, SF was underestimated almost 61% for
Moriasi et al. (2007), the model can be classified as satisfactory if the watersheds (observed SF of 109 and 430 dm3 s1, and simulated
NSE >0.5 and Pbias ± 25% for water yield and Pbias ± 55% for sedi- of 420 and 170 dm3 s1 for the NW and SW, respectively). SY
ment yield at monthly time scale. However, the validation showed simulation was 32% and 63% less than the observed SY for the NW
better performance than calibration due to the calibration data and SW, respectively. Overall, the SY calibration and validation for
(2016, 2017, and 2019) had dry and wet years (Fig. 2), becoming both watersheds were underestimated, indicated by the positive
hard to represent in the hydrological model. values of Pbias, and for SF calibration and validation were over-
Hydrographs and sedimentographs in Fig. 2 demonstrated that estimated (except in SW validation), indicated by the negative Pbias.
the performance of the model allowed the reproduction of the daily Furthermore, the baseflow simulation in the SW was overestimated
temporal variability in observed streamflow and sediment yield. in most of the observed period (Fig. 2b).
While the overall performance was satisfactory for streamflow and The difference between the statistics in the calibration and
sediment yield, several peaks were underestimated during the wet validation periods, especially for the North Watershed, can be
5
Ebling et al.
Table 3
Fitted values of the calibrated parameters for streamflow and sediment yield simulations in North (NW) and South (SW) watersheds.
Calibrated variable Parameters Name Fitted value
NW SW
Streamflow Baseflow alpha factor (days) v__ALPHA_BF.gw 0.09 0.26

Effective hydraulic conductivity in main channel alluvium (mm/hr) r__CH_K2.rte 0.01 e
Manning's "n" value for the main channel r__CH_N2.rte 0.03 e
SCS runoff curve number r__CN2.mgt 0.01 0.09
Soil evaporation compensation factor v__ESCO.hru 0.83 0.78
Groundwater delay time (days) v__GW_DELAY.gw 160.63 16.96
Threshold depth of water in the shallow aquifer required for return flow to occur (mm H2O) v__GWQMN.gw 2807.21 e
Groundwater "revap" coefficient v__GW_REVAP.gw 0.09
Average slope steepness (m/m) r__HRU_SLP.hru 0.07 e
Lateral flow travel time (days) v__LAT_TTIME.hru 2.75 1.64
Manning's "n" value for overland flow r__OV_N.hru 0.18 e
Deep aquifer percolation fraction v__RCHRG_DP e 0.06
Available water capacity of the soil layer (mm H2O/mm soil) r__SOL_AWC.sol 0.09 0.27
Moist bulk density (g/cm3) r__SOL_BD.sol 0.13 0.09
Saturated hydraulic conductivity (mm/hr) r__SOL_K.sol 0.25 e
Sediment yield Channel erodibility factor v__CH_COV1.rte 0.25 0.50

Peak rate adjustment factor for sediment routing in the main channel v__PRF_BSN.bsn 0.33 e
Linear parameter for maximum amount of sediment reentrained in channel sediment routing v__SPCON.bsn 0.01 0.01
Exponential coefficient for overland flow v__EROS_EXPO.bsn e 1.58
Rill erosion coefficient v__RILL_MULTI.bsn 1.02 1.03
Scaling parameter for cover and management factor for overland flow erosion v__C_FACTOR.bsn 0.06 0.03
USLE equation soil erodibility r__USLE_K.sol 0.18 e
Peak rate adjustment factor for sediment routing in the subbasin v__ADJ_PKR.bsn 1.30 e
Table 4 yr1 for the North watershed, and 5 dm3 s1 yr1 and 4.57 Mg ha1
Model performance statistics of streamflow and sediment yield in the calibration yr1 for the South watershed. The streamflow and sediment yield
(2016, 2018, and 2019) and validation (2017) periods in North (NW) and South (SW)
varied considerably in the different subbasins. Fig. 3 shows the
watersheds.
average annual water and sediment yield at each subbasin.
Analysed variable Simulation period NW SW Streamflow ranged from 3 (SB4) to 13.1 dm3 s1 yr1 (SW5), and
NSE R2 Pbias NSE R2 Pbias sediment yield was from 0.19 (SB6) to 89.7 Mg ha1 yr1 (SB5) for
Streamflow calibration 0.5 0.5 9.8 0.5 0.5 5.4
NW and between 0.6 (SB7) to 16.0 dm3 s1 yr1 (SB1), and 0.29
validation 0.9 1.0 24.4 0.6 0.7 23.2 (SB9) to 18.09 Mg ha1 yr1 (SB5) for SW, respectively.
Sediment yield calibration 0.5 0.5 35.6 0.4 0.4 22.3 Annual average streamflow (dm3 s1 yr1) and sediment yield
validation 0.9 0.9 3.3 0.4 0.4 30.6 (Mg ha1 yr1) from each subbasin were regrouped into different
Source data: Silva (2022). scales according to the behavior of both watersheds and to identify
the critical subbasins (Fig. 3). The obtained streamflow was cate-
gorized into four classes: 0e1 dm3 s1 yr1, 1e5 dm3 s1 yr1,
related to the small set of data available, for calibration 3 years were 5e10 dm3 s1 yr1, and >10 dm3 s1 yr1. The sediment yield was
used, and only 1 year for the validation period. The years reserved categorized into five classes: 0e5 Mg ha1 yr1, 5e15 Mg ha1 yr1,
for calibration had a dry year (2017) and a wet year (2019) as can be 15e50 Mg ha1 yr1, 50e80 Mg ha1 yr1, and >80 Mg ha1 yr1.
seen in Fig. 2, this poses a main difficulty for the representation in Most of subbasins from NW (85%) are under SF less than 5 dm3 s1
hydrological models, such as SWAT, that tend to the underestimate yr1, followed by 13.7% between 5 and 10 dm3 s1 yr1, and only
high peaks and super estimate low flows (Silva et al., 2023). While 1.3% of total area with SF more than 10 dm3 s1 yr1. The SW
for the validation period we had a milder year, and within the year, showed more streamflow compared to NW, in which the average SF
the peaks had a very good fit between the observed and simulated, of SW was 40% higher than the NW and larger area with SF more
as we know this probably can be due to good representation of than 10 dm3 s1 yr1. Therefore, 35.7% of total SW showed SF less
rainfall observation in the basin for these events, as for the cali- than 5 dm3 s1 yr1, 42.3% between 5 and 10 dm3 s1 yr1, and 22%
brated parameters, and this agreement improved especially the more than 10 dm3 s1 yr1. The annual average sediment yield for
NSE values, since it gives more emphasis on high flows. Therefore, NW and SW was 57.3 and 67.4% less than 5 Mg ha1 yr1, 36.3 and
the performance of the validation can be better than the calibra- 27% between 5 and 15 Mg ha1 yr1, 5.1 and 5.6% from 15 to
tion, Dakhlaoui et al. (2017) tested the how reliable and robust 50 Mg ha1 yr1, respectively. Only NW showed SY more than
hydrology models under different climate conditions and 80 Mg ha1 yr1, representing 1.3% of the total area.
concluded that the performance of the models can have different The most critical subbasins for streamflow and sediment yield
rates depending on the climate classes used for the calibration and were SB5 and SB7 for NW and SB1 and SB5 for SW, located near the
the validation periods and that the drier periods were harder to outlet of each watershed (Fig. 3a and b). Most of HRUs from the NW
address, as the year 2017 in our study. and SW’s critical subbasins belong to land cover type of cropland
(corn) and grassland (SW) associated with the higher slope (>8%)
and unpaved road (NW). Among subbasins with a lower sediment
3.2. Streamflow and sediment yield responses in paired watersheds
yield rate (less than 5 Mg ha1 yr1) are located Southeast of both
watersheds. The majority of these HRUs corresponds to a high
Based on simulated data, the average annual streamflow (SF)
percentage of native forest for NW and less slope for SW.
and sediment yield (SY) were 3 dm3 s1 yr1 and 11.20 Mg ha1
6
Ebling et al.
Fig. 2. Daily simulated and observed water yield (WY) and sediment yield (SY) calibration and validation in (a) North and (b) South watersheds using SWAT model.
3.3. Impact of BMPs at watershed scale and 6.0% for SOIL_BMP1, 0.4 and 0% for SOIL_BMP2, 8.7 and 6.7% for
SOIL_BMP3, 13.2 and 8.1% for VEG_BMP1, 38.4 and 28.8% for
Simulations of the nine BMPs showed the reduction of average VEG_BMP2, and 0 and 0.4% for VEG_BMP3, 1.6 and 9.1% for
annual sediment yield for North and South watersheds (Fig. 4a). STR_BMP1, 27.6 and 8.1% for STR_BMP2, and 20.4 and 13.9% for
Only BMPs of manure application (SOIL_BMP2) and grazing man- STR_BMP3, respectively. The most effective BMPs at each approach
agement (VEG_BMP3) did not affect the reduction of SY at water- were residue management (SOIL_BMP1) and conservation tillage
shed scale for SW and NW, respectively. The simulated average (SOIL_BMP3) for soil management; Crop rotation and cover crop
annual sediment yield at the base conditions was 11.20 Mg ha1 (VEG_BMP2) for vegetative measures; and contour farming
yr1 and 4.57 Mg ha1 yr1 for NW and SW, respectively. (STR_BMP2) and terraces (STR_BMP3) for structural practices.
After the implementation of residue management (SOIL_BMP1), Hence, the highest reduction efficiency was VEG_BMP2 for both
manure application (SOIL_BMP2), conservation tillage watersheds, followed by STR_BMP3 for NW and STR_BMP4 for SW.
(SOIL_BMP3), strip cropping (VEG_BMP1), crop rotation and cover
crop (VEG_BMP2), grazing management (VEG_BMP3), grassed 3.4. Impact of combined BMPs at watershed and subbasin scales
waterways (STR_BMP1), contour farming (STR_BMP2), and
terracing (STR_BMP3) provided the reduction of average annual After evaluating the impact of each BMP to reduce the average
sediment yield in both watersheds. annual sediment yield, four scenarios were built with combined of
Thereby, the North and South sediment yield has reduced by 9.3 the three approaches of BMPs: Scenario 1 (soil management BMPs:
7
Ebling et al.
Fig. 3. Average annual (a) water and (b) sediment yield at each sub-watershed under base scenario in North (NW) and South (SW) watersheds using SWAT model.
SOIL_BMP1, SOIL_BMP2, and SOIL_BMP3), Scenario 2 (vegetative to 11.87 (SB5) Mg ha1 yr1, respectively (Fig. 5b). The application
measures: VEG_BMP1, VEG_BMP2, and VEG_BMP3), Scenario 3 of this scenario affected the critical subbasins for both watersheds.
(structural practices: STR_BMP1, STR_BMP2, and STR_BMP3), and NW’s critical subbasins (SB5 and SB7) moved down from SY class
Scenario 4 (all BMPs, except VEG_BMP1). The average annual more than 80 Mg ha1 yr1 to class between 15 and 50 Mg ha1
sediment yield and reduction efficiency of the four scenarios were yr1, and the SB5 from SW shifted from SY class between 15 and
greater than individual BMPs (Fig. 4b), except for the combined 50 Mg ha1 yr1 to class 5e15 Mg ha1 yr1 (an increase of 6% of
structural practices in the NW. total area in this class). This scenario has also shifted SB14 from NW
The highest sediment reduction efficiency at watershed scale to class less than 5 Mg ha1 yr1 (increase 16.4% of total area in this
(46% and 41.5% for NW and SW, respectively) was achieved by the class), and the SB10 to class between 5 and 15 Mg ha1 yr1. The
implementation of combined all BMPs (Scenario 4) that provided implementation of the Vegetative scenario reduced SY between 10
the average annual sediment yield of 6.05 and 2.67 Mg ha1 yr1 for and 60%, with a reduction of 40e50%, representing 70% of total NW
NW and SW, respectively. Followed by the vegetative (Scenario 2), (including the critical subbasins), and a reduction from 30 to 40%,
structural (Scenario 3), and soil scenarios (Scenario 1) with a representing 80% of total SW (Fig. 6b).
reduction of 43.5 and 34.1%, 14.6 and 16%, and 9.9 and 6.8% for NW The structural scenario (Scenario 3) decreased the average
and SW compared to base scenario, respectively. annual SY in both watersheds that varied from 0.19 (SB6) to 76.46
Simulated annual average sediment yield from subbasins under (SB5) Mg ha1 yr1 for NW, and 0.27 (SB9) to 15.11 (SB5) Mg ha1
the four scenarios is shown in Fig. 5. Average annual of SY from Base yr1 for SW (Fig. 5c). The critical subbasins for NW (SB5 and SB7)
Scenario of NW and SW subbasins ranged from 0.19 (SB6) to shifted from the highest SY class (>80 Mg ha1 yr1) to class be-
89.73(SB5) Mg ha1 yr1 and 0.29 (SB9) to 18.09 (SB5) Mg ha1 tween 50 and 80 Mg ha1 yr1. SB2 and SB4 from NW have also
yr1, respectively (Fig. 3b). At the implementation of Scenario 1, the shifted to the soil loss class less than 5 Mg ha1 yr1. The imple-
average annual sediment yield ranged from 0.17 (SB6) to 81.11 (SB5) mentation of Scenario 3 did not modify SY classes of SW subbasins.
Mg ha1 yr1 for NW, and from 0.27 (SB9) to 16.97 (SB5) Mg ha1 However, the reduction of sediment yield ranged from 0 to 30% for
yr1 for SW (Fig. 5a). This scenario affected more the NW than SW, both watersheds, in which 65% of total area from NW reduced
in which the SY class from SW critical subbasins have not shifted. In between 10 and 20%, and 60% of SW area reduced from 0 to 10%
the application of Scenario 1 in NW, only SB7 moved down from (Fig. 6c).
class more than 80 Mg ha1 yr1 to class between 50 and Scenario 4 was the association of all BMPs types (soil, vegetative
80 Mg ha1 yr1. This scenario has also moved down SB14 of NW to and structural measures), and this showed the most effective to
class less than 5 Mg ha1 yr1 (increase 6% of total area in this decrease sediment yield in both watersheds (Figs. 5d and 6d). The
class). The sediment yield with the implementation of Scenario 1 average annual SY ranged from 0.14 (SB6) to 48.38 (SB5) Mg ha1
has reduced from 0 to 10% in critical subbasins, and 10e30% in the yr1 for NW and 0.19 (SB6) to 10.26 (SB5) Mg ha1 yr1 for SW
other subbasins, with the highest SY reduction (20e30%) detected (Fig. 5d). Compared to base scenario, the subbasins from NW
in SB16 from NW (Fig. 6a). shifted as the vegetative scenario (Scenario 2). However, SB1 and
The average annual sediment yield of NW and SB under Scenario SB5 from SW moved down to SY class less than 5 (increased 16% of
2 ranged from 0.15 (SB6) to 48.88 (SB5) Mg ha1 yr1 and 0.19 (SB9) total SW) and between 5 and 15 Mg ha1 yr1, respectively.
8
Ebling et al.
Fig. 4. Sediment yield reductions (%) at (a) individual and (b) combined BMPs (Scenarios) compared to the Base scenario in North (NW) and South (SW) watersheds simulated by
SWAT model.
Reduction of SY varied from 20 to 60% in this scenario, in which 96% 4. Discussion

of total area from NW reduced between 40 and 60%, and 90% of SW
total area from 30 to 50% (Fig. 6d). 4.1. Soil erosion dynamics in the paired watersheds
Studied paired watersheds consist predominantly of soil classes

3.5. Impact of combined BMPs on water balance components that are less susceptible to soil erosion, but are associated with high
rainfall, steep slope, and cultivated land, which enable soil erosion
Average annual water balance components have been estimated processes. SWAT model was used after daily streamflow and sedi-
(2016e2019) for the base scenario and the four combined BMPs ment yield calibration and validation to evaluate the effectiveness
scenarios at each watershed (Fig. 7). In the base scenario, evapo- of individual and combined BMPs on sediment yield in two agri-
transpiration was more predominant in the NW and SW which cultural paired watersheds. Firstly, based on simulated and
accounted for 64.5 and 49.5% of the average annual rainfall (1122.3 observed data in four years (2016e2019), NW showed more sus-
and 1512.9 mm), respectively. From the rainfall generated flow, 12.4 ceptibility to soil erosion than SW. The average annual sediment
and 19.6% were as surface runoff, and 7.1 and 28.5% were as base- yield in NW (11.20 Mg ha1 yr1) was almost 60% greater than in
flow for NW and SW, respectively. Results of the implementation of SW (4.57 Mg ha1 yr1). These watersheds are different mainly in
four scenarios indicated a marginal change in the annual average size of drainage area and the presence of riparian vegetation
surface runoff, total aquifer recharge, percolation, evapotranspira- (Ebling, 2018). Some studies have reported the presence of riparian
tion, and baseflow for both watersheds (Fig. 7). There was a vegetation along the drainage network decreases the amount of
reduction in the surface runoff in all scenarios for SW and NW sediment transported and mobilized to the water bodies
compared to Base scenario, ranging from 14.8 to 6.4%. (Sirabahenda et al., 2020; Tiecher et al., 2017; Waidler et al., 2011).
The reduction in the surface runoff could be supported by an Riparian vegetation could work as a physical filter that reduces
increase in total aquifer recharge (3.1e14.8% for NW, 2.7e5.8% for excessive amounts of sediment, nutrients, and pesticides in surface
SW), an increase of percolation (3.8e17.3% for NW, 2.6e5.5% for runoff (Broetto et al., 2017; Waidler et al., 2011). Besides, riparian
SW) and an increase of baseflow (2.9e13.86% for NW, 2.7e5.9% for vegetation decreases connectivity between cropland and streams
SW), and a decrease and increase in evapotranspiration (5.2 to (Tiecher et al., 2017).
0.2% NW, 0.1e2.1% for SW). The impact in water balance compo- Critical subbasins from NW and SW showed sediment yields of
nents by Scenario 4 was higher than the other three scenarios in more than 80 Mg ha1 yr1 and more than 15 Mg ha1 yr1,
both watersheds, followed by Scenario 2.
9
Ebling et al.
Fig. 5. Mean annual sediment yield at each sub-basin under (a) Scenario 1, (b) Scenario 2, (c) Scenario 3, and (d) Scenario 4 in North and South watersheds.
respectively. According to the soil erosion class developed for et al., 2020). Although unpaved road covers a small portion of the
Brazilian conditions by Carvalho (2008), these subbasins can be NW area, these have shown a significant source of runoff and
classified by severe and moderate soil loss. These subbasins were sediment in this study, due to unpaved roads receives the stream-
attributed mainly to steep slopes, higher percent of cropland, and flow and sediments from cultivated slopes. The same was observed
grassland associated with the presence of unpaved roads in NW. by Minella et al. (2007, 2009), the unpaved roads occupied a small
There is a significant association between slope gradient and portion of the watershed but resulted in a large contribution to soil
land use with soil loss. Some studies argued there is a linear effect erosion. Thomaz et al. (2014) showed that drainage areas less than
of slope gradient with soil loss increase (Rieke-Zapp & Nearing, 3 km2 had more contribution to sediment yield than large areas.
2005; Zhang et al., 2021), in which steeper slope gradients tend Unpaved roads systems may change the hydrologic surface by
to decrease soil infiltration and increase the velocity of surface increasing the concentrated runoff (Wang, Wu, et al., 2021) and
runoff, promoting greater erosion and sediment transport (Deng consequently contributing to more sediment yield in these areas.
10
Ebling et al.
Fig. 6. Reduction in sediment yield (%) at each sub-basin after implementation of (a) Scenario 1, (b) Scenario 2, (c) Scenario 3, and (d) Scenario 4 in North and South watershed
compared to Base Scenario.
However, the introduction of improved management practices in erosion (Blanco-Sepúlveda et al., 2021). Currently, the no-till in
the fields can reduce sediment yield from unpaved roads. As noted both watersheds has been implemented but without the premises
by Minella et al. (2009), the implementation of minimum tillage of the no-till system, only keep the ground cover from previous
reduced runoff on fields and, consequently, reduced runoff and soil crops, resulting in a high streamflow and sediment yield. Thus, it is
erosion onto roads. essential to implement conservation practices in these areas.
Previous studies have shown that land use is the key factor that
affects soil erosion (Anache et al., 2017) due to minimize the effect
of rainfall splashes and surface runoff. For example, under the same 4.2. Impact of individual and combined BMPs in the paired
rainfall amount, the sediment yield in bare soil and cropland was watersheds
significantly higher than in natural vegetation (Zhang et al., 2021).
Cropland areas are the main sediment source in most watershed The effectiveness of sediment yield reduction of individual
studies (Risal et al., 2020; Tiecher et al., 2015, 2018). Tillage and BMPs varied from 0 to 38.4% for NW and 0e28.8% for SW. The most
management in these areas induce soil erosion (Zhao et al., 2018), effective conservation practice in both watersheds was crop rota-
while adoption of conservation tillage is effective to reduce soil tion and cover crop (VEG_BMP2). The VEG_BMP2 is an effective
conservation practice to mitigate soil erosion due to higher soil
11
Ebling et al.
Fig. 7. Change in water balance components after implementation the BMPs scenarios compared to Base Scenario in (a) North and (b) South watersheds.
cover and roughness resulting from different vegetations (Didone Among the three approaches with soil, vegetative and structural
et al., 2017). The NW and SW are characterized by intensive soy- conservation measures, the combined vegetative measures (Sce-
bean/corn monoculture which limits the effects of crops diversity. nario 2) were the most effective to reduce sediment yield at
For example, da Silva et al. (2021) showed that soybean mono- watershed (reduction of 43.5% for NW and 34.1% for SW) and
culture had similar soil, water, and nutrient losses as bare soil. subbasin (reduction from 10 to 60% for both watersheds) scale. The
Therefore, the implementation of crop rotation and cover crop combination of both conservation practices reduces the effect of
enhances soil environmental and agronomic functions by raindrop impact by cover crop, besides reducing the velocity of
increasing rainfall intercept, which minimizes the direct impact of surface runoff from strip cropping and soil-transporting capacity,
raindrops on the soil surface, and decreases surface sealing (Blanco- and, consequently, decreases soil loss (Laufer et al., 2016;
Canqui & Ruis, 2018). Wischmeier & Smith, 1978). Laufer et al. (2016) showed that strip
The second most effective individual BMPs were contour cropping with crop rotation could reduce 92 and 98% of surface
farming for NW and terraces for SW that SY reduced 27.6 and 13.9%, runoff and soil loss compared to intensive tillage, respectively.
respectively. Both BMPs were considered structural practices in this Structural practices (Scenario 3) were the second combined
study, in which the purpose is to reduce the velocity and volume of BMP to sediment yield reduction. Scenario 3 could reduce from 0 to
surface runoff. Terraces can substantially reduce runoff, especially 30% of sediment yield in subbasins for both watersheds. In the
during heavy rainfall events. Ran et al. (2020) showed that under a combined structural BMPs, the reduction of sediment yield is
rainfall intensity of 120 mm h1, a well-maintained terrace could caused by the reduced velocity, volume, peak, and erosive power of
reduce runoff by 100% compared to a natural hillslope. However, surface runoff through impounding water in small depressions and
the major limitation of terracing is caused by poor management, reduced length of hillslope (Arabi et al., 2008; Chen et al., 2012;
which could increase soil loss from 1 to 5 times than well- Huihui et al., 2016). Combined structural BMPs decreased the
management terraces (Deng et al., 2021). In contrast, in some average annual sediment yield at watershed scale and in the critical
et al., 2021; Karlen et al., 2009),
studies (Briak et al., 2019; Didone subbasins from NW. Contour farming is most effective on gentle
the implementation of contour farming was highly effective in and shorter slopes (Jia et al., 2020; USDA, 2017) that longer and
controlling erosive processes in both watersheds. Dibaba et al. steeper slopes, overland flow volume and velocity exceed the ca-
(2021) also had a high reduction of contour farming, but terracing pacity of the contour ridges. Therefore, increasing roughness by
resulted in a higher reduction of soil loss than contour farming. implementing terraces decrease surface runoff and sediment yield
12
Ebling et al.
(Fang, 2021; USDA, 2017). (Reichert et al., 2009, 2017, 2020; Vaz et al., 2005). Improving water
The least efficient scenario between the three approaches was infiltration is a key to keeping water available, especially in dry
soil management. The benefits of the soil management scenario are periods. In dry periods, the surface water in watercourses is
to reduce surface runoff by increasing land cover and roughness, recharged by groundwater (Fan et al., 2013). Thereby, the increase
improve soil aggregate stability, increase infiltration, and then, of baseflow and total aquifer recharge and the contribution of
decrease soil loss (Arabi et al., 2008). Some studies observed that groundwater to surface water by conservation practices could be an
the positive effects on physical properties and on soil erosion have interesting option for farmers (Boufala et al., 2021).
shown to a long term in implementing these measures (Klik & With the implementation of Scenario 2 and 4, we could observe
Rosner, 2020; Nunes et al., 2018; Sithole et al., 2019; Wolschick a small decrease and an increase in evapotranspiration for NW and
et al., 2021; Zanon et al., 2020; Zhang et al., 2007). SW, respectively. The change in the diversity of vegetations and
The adoption of no-till in Brazil has increased in the last decade, increase the soil cover by crop cover could provide lower evapo-
Fuentes-Llanillo et al. (2021) observed an increase of 84.9% in no-till transpiration in NW. Yang et al. (2018) observed that maize-wheat-
areas between 2006 and 2017. However, the application of only no- soybean rotation under no-till provided higher soil water storage
till is not guaranteed for production sustainability and optimiza- and lower evapotranspiration compared to conventional tillage. But
tion. Zanon et al. (2020) studied the long-term (twenty years) effect in only maize land, transpiration and evaporation were not signif-
on manure application in no-till areas, and observed the icantly changed by the different treatments. The same was found by
improvement in physical, chemical, and biological properties. Boufala et al. (2021), no effect on evapotranspiration was observed
However, it was not enough to reduce runoff under high-intensity in the three BMPs scenarios. In general, Scenario 2 and 4 (Figs. 6 and
rainfall, even with the presence of straw and absence of surface 7) were the most appropriate for the integrated management for
sealing. Because of this, the association of different types of con- both watersheds, which provide a better result to minimize the
servation practices is necessary to minimize the effects of soil erosion processes and change the water balance in watersheds.
erosion.
The most efficient scenario included all types of BMPs (Scenario
4). Scenario 4 reduced sediment yield of 46 and 41.5% for NW and 4.4. Strengths and limitations of the study
SW at watershed scales from base scenario, respectively. From 20 to
60% at subbasin scales, which reduced from 40 to 50% in the critical Findings of this study will contribute to helping decision-
subbasins. Previous studies have reported that the association of makers, farmers, and water resources planners, as it provides in-
vegetative and structural conservation measures is the best way to formation about the most susceptible areas for soil erosion and the
control soil erosion (Ebabu et al., 2019; Gashaw et al., 2021; Lo pez- implementation of the best management practices to minimize the
Ballesteros et al., 2019). For example, Uniyal et al. (2020) indicated effects of soil erosion and hydrological processes. One of the lessons
the good performance of the combined agronomic and structural learned from this research is that the implementation of only "no-
BMPs in controlling sediment yield than individual BMPs in the till" is not enough to contain soil erosion, resulting in a high sedi-
Baitarani watershed (India). Lo pes-Ballesteros et al. (2019) associ- ment yield rate in both watersheds. However, the adoption of BMPs
ated five structural and agricultural BMPs (reforestation, check dam resulted in soil erosion control for most of the tested BMPs. The
restoration, contouring, filter strip, and fertilizer application), and most effective individual BMP for both watersheds was VEG_BMP2
found reduced values of sediment yield until 93%, compared to (crop rotation and cover crop), indicating that the good manage-
scenario without BMP. In general, structural practices are designed ment with diverse vegetive crops and to keep surface soil protected
to control soil erosion and surface runoff where soil management could decrease the effects of soil erosion and improve the soil
and vegetative practices are projected to improve soil quality and conditions. Besides being a conservation practice of easy imple-
decrease the impact of raindrops (Morgan, 2009), but these prac- mentation for farmers, it has high economic benefits. Another
tices alone are insufficient to reduce soil erosion to permissible lesson learned was about the different types of BMPs, in which the
levels (Blanco & Lal, 2010). association of soil, vegetative and structural conservation measures
resulted in the best management to decrease soil erosion. For
4.3. Impact of combined BMPs on water balance components example, Scenario 4 was an effective scenario to reduce soil erosion
and change water balance components.
There was a great impact of combined BMPs on water balance Based on the outcomes of this study, there are some limitations.
components (surface runoff, total aquifer recharge, percolation, First of all, there was an uncertainty associated with modeling due
evapotranspiration, and baseflow). Scenario 4 had more impact on to several factors including lack of hydrossedimentological vari-
water balance components, followed by Scenario 2. In both wa- ables observations and measurements, and limitations to represent
tersheds, there was a reduction of surface runoff, an increase in the soil erosion and hydrological processes in the SWAT model.
total aquifer recharge, percolation, evapotranspiration, and base- Secondly, the streamflow and sediment yield modeling period were
flow. This result was similar to Uniyal et al. (2020) that the small for calibration and validation. Therefore, increasing this
implementation of three scenarios resulted in a reduction of sur- period with the monitored data could better represent the soil
face runoff, and then an increase of lateral flow, aquifer recharge, erosion and hydrological processes.
baseflow, and percolation. Base scenario indicated that 12.4 and Finally, this study showed the efficiency of BMPs implementa-
19.6% of total rainfall was generated in the surface runoff for NW tion in reducing sediment yield and optimizing water balance, but
and SW, respectively. the economic feasibility of these implementations was not evalu-
Implementation of all scenarios decreased the surface runoff ated. Economic feasibility by BMPs is an important subject to
volume and velocity by increasing and improving in-watershed inform and convince farmers to adopt BMPs in croplands. In gen-
utilization of water, such as increasing infiltration rate, and in eral, several barriers limit the adoption of BMPs in many countries.
turn minimizing soil erosion (Himanshu et al., 2019). Both water- Therefore, programs for soil and water conservation must be
sheds are characterized by predominantly clay soils (Ferralsols and implemented by the public and private sectors to provide subsidies
Nitisols) with a moderate soil permeability (Holthusen, Brandt, for farmers (Brazilian programs: “Water Conservation”, “Forest
Reichert, Horn, et al., 2018; Holthusen, Brandt, Reichert, & Horn, Assistance Program”, “Conserv”, “Reforest Program”, “Water
2018; Mentges et al., 2016) but high-water retention capacity Producer”).
13
Ebling et al.
5. Conclusions Arabi, M., Frankenberger, J. R., Engel, B. A., & Arnold, J. G. (2008). Representation of
agricultural conservation practices with SWAT. Hydrological Processes: Interna-
tional Journal, 22(16), 3042e3055.
In this study, monitoring and modeling of two paired agricul- Arnold, J., Srinivason, R., Muttiah, R., & Williams, J. (1998). Large area hydrologic
tural watersheds were carried out using SWAT model to evaluate modeling and assessment. Part I: Model development. Journal of the American
the effectiveness of nine potential soil, vegetative and structural Water Resources Association, 34(1), 73e89.
ASF. (2019). Data search of the DEM. Alaska Satellite Facility (ASF). Retrieved from
best management practices (BMPs) to reduce sediment yield. Very https://search.asf.alaska.edu/. No pagination.
high sediment yields were measured in both watersheds, in which Berihun, M. L., Tsunekawa, A., Haregeweyn, N., Dile, Y. T., Tsubo, M., Fenta, A. A., …
North watershed (NW) showed more susceptibility to soil erosion Srinivasan, R. (2020). Evaluating runoff and sediment responses to soil and
water conservation practices by employing alternative modeling approaches.
than South watershed (SW). These results illustrate that the pres- Science of the Total Environment, 747, Article 141118.
ence of riparian vegetation in SW helps to reduce sediment yield, Bertoni, J., & Lombardi Neto, F. (2014). Conservaça ~o do solo. Sa~o Paulo: Icone.
and the no-till alone cannot hold soil erosion. About the BMPs Blanco-Canqui, H., & Ruis, S. J. (2018). No-tillage and soil physical environment.
Geoderma, 326, 164e200.
simulations, VEG_BMP2 (crop rotation and cover crop) was the Blanco-Sepúlveda, R., Enríquez-Narva ez, F., & Lima, F. (2021). Effectiveness of con-
most effective individual BMP for both watersheds. The combina- servation agriculture (tillage vs. vegetal soil cover) to reduce water erosion in
tion of all BMPs approaches (Scenario 4) showed the most effective maize cultivation (Zea mays L.): An experimental study in the sub-humid up-
lands of Guatemala. Geoderma, 404, Article 115336.
to reduce sediment yield (46 and 41.5% for NW and SW, respec- Blanco, H., & Lal, R. (2010). Soil and water conservation. In Principles of soil con-
tively). Followed by the vegetative combination (Scenario 2) that servation and management (p. 2). Springer.
reduced SY by 43.5 and 34.1% for NW and SW, respectively. Sce- Borrelli, P., Robinson, D. A., Fleischer, L. R., Lugato, E., Ballabio, C., Alewell, C., …
Panagos, P. (2017). An assessment of the global impact of 21st century land use
narios 2 and 4 could reduce from 40 to 50% of sediment yield in the
change on soil erosion. Nature Communications, 8(1), 1e13.
critical subbasins. In cases where applying all measures is not Boufala, M. H., El Hmaidi, A., Essahlaoui, A., Chadli, K., El Ouali, A., & Lahjouj, A.
possible, Scenario 2 is a potential scenario that can be used to (2021). Assessment of the best management practices under a semi-arid basin
minimize soil erosion. The implementation of combined scenarios using SWAT model (case of M’dez watershed, Morocco). Modeling Earth Systems
and Environment, 1e19.
supported improving water balance components, which decreased Bressiani, D. A. (2016). Coping with hydrological risks through flooding risk index,
surface runoff from 6.4 to 14.8%, and increased groundwater com- complex watershed modeling, different calibration techniques, and ensemble
streamflow forecasting (Doctoral dissertation). Sa ~o Carlos, SP: Universidade de
ponents from 2.7 to 18.9% in both watersheds.
Sa~o Paulo.
The results provide some evidences that the implementation of Bressiani, D. A., Gassman, P. W., Fernandes, J. G., Garbossa, L. H. P., Srinivasan, R.,
only one type of conservation measures approach is not enough to Bonuma , N. B., & Mendiondo, E. M. (2015). Review of soil and water assessment
reduce soil loss, and the association of the three different BMPs tool (SWAT) applications in Brazil: Challenges and prospects. International
Journal of Agriculture and Biology, 8, 9e35.
approaches minimizes soil erosion processes and changes water Briak, H., Mrabet, R., Moussadek, R., & Aboumaria, K. (2019). Use of a calibrated
balance. However, more studies are needed to assess the impacts of SWAT model to evaluate the effects of agricultural BMPs on sediments of the
BMP implementation on crop productivity, cost benefits, and in Kalaya river basin (North of Morocco). International Soil and Water Conservation
Research, 7(2), 176e183.
high erosivity and water stress conditions. Nonetheless, the find- Broetto, T., Tornquist, C. G., Campos, B. H. C. D., & Schneider, J. C. (2017). Relation-
ings of this study could help farmers and decision-makers choosing ships between agriculture, riparian vegetation, and surface water quality in
feasible and appropriate conservation measures to reduce the watersheds. Revista Brasileira de Ci^ encia do Solo, 41, Article e0160248.
Carvalho, N. O. (2008). Hidrossedimentologia pra tica. 2a ediça
~o, revisada e ampliada.
impact of erosive processes. ^ncia.
Rio de Janeiro: Intercie
Chen, S. K., Liu, C. W., & Chen, Y. R. (2012). Assessing soil erosion in a terraced paddy
Declaration of competing interest field using experimental measurements and Universal Soil Loss Equation. Ca-
tena, 95, 131e141.
Dakhlaoui, H., Ruelland, D., Tramblay, Y., & Bargaoui, Z. (2017). Evaluating the
The authors declare that they have no known competing robustness of conceptual rainfall-runoff models under climate variability in
financial interests or personal relationships that could have northern Tunisia. Journal of hydrology, 550, 201e217.
Deng, L., Sun, T., Fei, K., Zhang, L., Fan, X., Wu, Y., & Ni, L. (2020). Effects of erosion
appeared to influence the work reported in this paper. degree, rainfall intensity and slope gradient on runoff and sediment yield for
the bare soils from the weathered granite slopes of SE China. Geomorphology,
352, Article 106997.
Acknowledgement
Deng, C., Zhang, G., Liu, Y., Nie, X., Li, Z., Liu, J., & Zhu, D. (2021). Advantages and
disadvantages of terracing: A comprehensive review. International Soil and
This study was financed in part by the “Coordenaça ~o de Aper- Water Conservation Research, 9, 344e359.
Dibaba, W. T., Demissie, T. A., & Miegel, K. (2021). Prioritization of sub-watersheds
feiçoamento de Pessoal de Nível Superior” (Capes) - Finance Code
to sediment yield and evaluation of best management practices in highland
001, Brazilian Council for Scientific and Technological Development Ethiopia, finchaa catchment. Land, 10(6), 650.
(CNPq), and Fundaça ~o de Amparo a Pesquisa do Estado do Rio Didone , E. J., Minella, J. P. G., & Evrard, O. (2017). Measuring and modelling soil
Grande do Sul (Fapergs). We thank Dr. Jose Eloir Denardin, from erosion and sediment yields in a large cultivated catchment under no-till of
South Brazil. Soil and Tillage Research, 174, 24e33.
EMBRAPA-Wheat, for providing information about the first stage of Didone , E. J., Minella, J. P. G., & Piccilli, D. G. A. (2021). How to model the effect of
the project. mechanical erosion control practices at a catchment scale? International Soil
and Water Conservation Research, 9(3), 370e380.
Didone , E. J., Minella, J. P. G., Schneider, F. J. A., Londero, A. L., Lefevre, I., & Evrard, O.
References (2019). Quantifying the impact of no-tillage on soil redistribution in a cultivated
catchment of Southern Brazil (1964e2016) with 137Cs inventory measure-
Abbaspour, K. C. (2008). SWAT calibration and uncertainty programs. A User Manual. ments. Agriculture, Ecosystems & Environment, 284, 106588.
Switzerland: Eawag Zurich. Ebabu, K., Tsunekawa, A., Haregeweyn, N., Adgo, E., Meshesha, D. T., Aklog, D., …
Abbaspour, K. C., Yang, J., Maximov, I., Siber, R., Bogner, K., Mieleitner, J., … Yibeltal, M. (2019). Effects of land use and sustainable land management
Srinivasan, R. (2007). Modelling hydrology and water quality in the pre-alpine/ practices on runoff and soil loss in the Upper Blue Nile basin, Ethiopia. Science of
alpine Thur watershed using SWAT. Journal of hydrology, 333(2e4), 413e430. the Total Environment, 648, 1462e1475.
Afroz, M. D., Li, R., Muhammed, K., Anandhi, A., & Chen, G. (2021). Best management Ebling, E. D. (2018). Dina ^mica hidrossedimentolo gica de duas bacias hidrogra ficas com
practices for sustaining agricultural production at choctawhatchee watershed produça ~o leiteira e gra ~os (Master thesis). Santa Maria, RS: Universidade Federal
in Alabama, USA, in Response to climate change. Air, Soil and Water Research, 14, de Santa Maria.
1e12. D., Reichert, J. M., Minella, J. P. G., Holthusen, D., Broetto, T., & Srinivasan, R.
Ebling, E.
Alvares, C. A., Stape, J. L., Sentelhas, P. C., De Moraes Gonçalves, J. L., & Sparovek, G. (2022). Rainfall event-based surface runoff and erosion in small watersheds
(2013). Ko €ppen’s climate classification map for Brazil. Meteorologische Zeits- under dairy and direct-seeding grain production. Hydrological Processes, 36,
chrift, 22(6), 711e728. Article e14688.
Anache, J. A., Wendland, E. C., Oliveira, P. T., Flanagan, D. C., & Nearing, M. A. (2017). Fan, Y., Chen, Y., Liu, Y., & Li, W. (2013). Variation of baseflows in the headstreams of
Runoff and soil erosion plot-scale studies under natural rainfall: A meta- the tarim river basin during 1960e2007. Journal of Hydrology, 487, 98e108.
analysis of the Brazilian experience. Catena, 152(2017), 29e39. Fang, H. (2021). Effect of soil conservation measures and slope on runoff, soil, TN,
14
Ebling et al.
and TP losses from cultivated lands in North China. Ecological Indicators, 126, Veith, T. L. (2007). Model evaluation guidelines for systematic quantification of
Article 107677. accuracy in watershed simulations. Transactions of the ASABE, 50(3), 885e900.
FAO, Food and Agricultural Organization of the United Nations. (2019). 5 reasons Nunes, M. R., van Es, H. M., Schindelbeck, R., Ristow, A. J., & Ryan, M. (2018). No-till
why we need to protect our soils. Retrieved from https://www.fao.org/fao- and cropping system diversification improve soil health and crop yield. Geo-
stories/article/en/c/1192794/. derma, 328, 30e43.
Freitas, L. D. R., Gubiani, P. I., Mulazzani, R. P., Minella, J. P. G., & Londero, A. L. (2021). Oliveira, P. T. S., Wendland, E., & Nearing, M. A. (2013). Rainfall erosivity in Brazil: A
Terracing increases soil available water to plants in no-tillage. Revista Brasileira review. Catena, 100, 139e147.
de Ciência do Solo, 45. Peixoto, D. W. B., Guasselli, L. A., & Filho, W. P. (2017). Influe ^ncia da precipitaça ~o
Fuentes-Llanillo, R., Telles, T. S., Junior, D. S., de Melo, T. R., Friedrich, T., & Kassam, A. pluviome trica nos valores de reflecta ^ncia da agua, no reservato rio Passo Real -
(2021). Expansion of no-tillage practice in conservation agriculture in Brazil. RS. Revista Brasileira de Cartografia, 69(3), 495e503.
Soil and Tillage Research, 208, Article 104877. Pimentel, D. (2006). Soil erosion: A food and environmental threat. Environment,
Gashaw, T., Dile, Y. T., Worqlul, A. W., Bantider, A., Zeleke, G., Bewket, W., & Development and Sustainability, 8(1), 119e137.
Alamirew, T. (2021). Evaluating the effectiveness of best management practices Priestley, C. H. B., & Taylor, R. J. (1972). On the assessment of surface heat flux and
on soil erosion reduction using the SWAT model: For the case of gumara evaporation using large-scale parameters. Monthly Weather Review, 100, 81e92.
watershed, abbay (upper blue nile) basin. Environmental Management, 68, Ran, Q., Chen, X., Hong, Y., Ye, S., & Gao, J. (2020). Impacts of terracing on hydro-
240e261. logical processes: A case study from the loess plateau of China. Journal of Hy-
Gassman, P. W., Sadeghi, A. M., & Srinivasan, R. (2014). Applications of the SWAT drology, 588, Article 125045.
model special section: Overview and insights. Journal of Environmental Quality, Reichert, J. M., Albuquerque, J. A., Kaiser, D. R., Reinert, D. J., Urach, F. L., &
43, 1e8. Carlesso, R. (2009). Estimation of water retention and availability in soils of Rio
Guerra, A. J. T., Fullen, M. A., Jorge, M. D. C. O., & Alexandre, S. T. (2014). Soil erosion Grande do Sul. Revista Brasileira de Ci^ encia do Solo, 33, 1547e1560.
and conservation in Brazil. Anuario do Instituto de Geociencias, 37(1), 81e91. Reichert, J. M., Albuquerque, J. A., Peraza, J. E. S., & da Costa, A. (2020). Estimating
Himanshu, S. K., Pandey, A., Yadav, B., & Gupta, A. (2019). Evaluation of best man- water retention and availability in cultivated soils of South Brazil. Geoderma
agement practices for sediment and nutrient loss control using SWAT model. Regional, 21, Article e00277.
Soil and Tillage Research, 192, 42e58. Reichert, J. M., Rodrigues, M. F., Pela ez, J. J. Z., Lanza, R., Minella, J. P. G., Arnold, J. G.,
Holthusen, D., Brandt, A. A., Reichert, J. M., & Horn, R. (2018). Soil porosity, & Cavalcante, R. B. L. (2017). Water balance in paired watersheds with euca-
permeability and static and dynamic strength parameters under native forest/ lyptus and degraded grassland in Pampa biome. Agricultural and Forest Mete-
grassland compared to no-tillage cropping. Soil and Tillage Research, 177, orology, 237, 282e295.
113e124. Reicosky, D. C. (2015). Conservation tillage is not conservation agriculture. Journal of
Holthusen, D., Brandt, A. A., Reichert, J. M., Horn, R., Fleige, H., & Zink, A. (2018). Soil Soil and Water Conservation, 70(5), 103Ae108A.
functions and in situ stress distribution in subtropical soils as affected by land Ricci, G. F., Jeong, J., De Girolamo, A. M., & Gentile, F. (2020). Effectiveness and
use, vehicle type, tire inflation pressure and plant residue removal. Soil and feasibility of different management practices to reduce soil erosion in an
Tillage Research, 184, 78e92. agricultural watershed. Land Use Policy, 90, Article 104306.
Huihui, Z., Na, T., & Qingfeng, Z. (2016). Spatial heterogeneity of loess contour tilled Rieke-Zapp, D. H., & Nearing, M. A. (2005). Slope shape effects on erosion: A lab-
microtopographic slope in rainfall erosion. Soil Science & Plant Nutrition, oratory study. Soil Science Society of America Journal, 69(5), 1463e1471.
62(5e6), 409e415. Risal, A., Parajuli, P. B., Dash, P., Ouyang, Y., & Linhoss, A. (2020). Sensitivity of hy-
Hu, J., Wu, Y., Wang, L., Sun, P., Zhao, F., Jin, Z., Wang, Y., Qiu, L., & Lian, Y. (2021). drology and water quality to variation in land use and land cover data. Agri-
Impacts of land-use conversions on the water cycle in a typical watershed in the cultural Water Management, 241, Article 106366.
southern Chinese Loess Plateau. Journal of Hydrology, 593, Article 125741. da Silva, L. C. M., Avanzi, J. C., Peixoto, D. S., Merlo, M. N., Borghi, E., Resende, A. V., &
Jia, L., Zhao, W., Zhai, R., An, Y., & Pereira, P. (2020). Quantifying the effects of Silva, B. M. (2021). Ecological intensification of cropping systems enhances soil
contour tillage in controlling water erosion in China: A meta-analysis. Catena, functions, mitigates soil erosion, and promotes crop resilience to dry spells in
195, Article 104829. the Brazilian Cerrado. International Soil and Water Conservation Research, 9,
Karlen, D. L., Dinnes, D. L., Tomer, M. D., Meek, D. W., Cambardella, C. A., & 591e604.
Moorman, T. B. (2009). Is no-tillage enough? A field-scale watershed assess- Silva, T. P. (2022). Modelagem do impacto do uso e manejo do solo nos processos
ment of conservation effects. Electronic Journal of Integrative Biosciences, 7(2), hidrolo gicos e erosivos em pequenas bacias hidrogra ficas (Doctoral dissertation.
1e24. Universidade Federal de Santa Maria).
Klik, A., & Rosner, J. (2020). Long-term experience with conservation tillage prac- Silva, T. P., Bressiani, D., Ebling, E. D., Deus Júnior, J. C., & Reichert, J. M. (2023).
tices in Austria: Impacts on soil erosion processes. Soil and Tillage Research, 203, Evaluating hydrological and soil erosion processes in diferente time scales and
Article 104669. land uses in Southern Brazilian paired watersheds. Hydrological Sciences Journal.
Landers, J. N. (2005). Histo rico, característica e benefícios do plantio direto. Brasília: https://doi.org/10.1080/02626667.2023.2217333
ABEAS. Sirabahenda, Z., St-Hilaire, A., Courtenay, S. C., & Van Den Heuvel, M. R. (2020).
€rla
Laufer, D., Loibl, B., Ma €nder, B., & Koch, H. J. (2016). Soil erosion and surface runoff Assessment of the effective width of riparian buffer strips to reduce suspended
under strip tillage for sugar beet (Beta vulgaris L.) in Central Europe. Soil and sediment in an agricultural landscape using ANFIS and SWAT models. Catena,
Tillage Research, 162, 1e7. 195, Article 104762.
Li, H., Wang, W., Fu, J., Chen, Z., Ning, Z., & Liu, Y. (2021). Quantifying the relative Sithole, N. J., Magwaza, L. S., & Thibaud, G. R. (2019). Long-term impact of no-till
contribution of climate variability and human activities impacts on baseflow conservation agriculture and N-fertilizer on soil aggregate stability, infiltra-
dynamics in the Tarim River Basin, Northwest China. Journal of Hydrology: tion and distribution of C in different size fractions. Soil and Tillage Research,
Regional Studies, 36, Article 100853. 190, 147e156.
Londero, A. L., Minella, J. P., Deuschle, D., Schneider, F. J., Boeni, M., & Merten, G. H. Strauch, M., Lima, J. E., Volk, M., Lorz, C., & Makeschin, F. (2013). The impact of Best
(2018). Impact of broad-based terraces on water and sediment losses in no-till Management Practices on simulated streamflow and sediment load in a Central
(paired zero-order) catchments in South Brazil. Journal of Soils and Sediments, Brazilian catchment. Journal of Environmental Management, 127, S24eS36.
18(3), 1159e1175. Sun, D., Yang, H., Guan, D., Yang, M., Wu, J., Yuan, F., Jin, C., Wang, A., & Zhang, Y.
Lopes, T. R., Zolin, C. A., Mingoti, R., Vendrusculo, L. G., de Almeida, F. T., de (2018). The effects of land use change on soil infiltration capacity in China: A
Souza, A. P., Oliveira, R. F., Paulino, J., & Uliana, E. M. (2021). Hydrological meta-analysis. Science of the Total Environment, 626, 1394e1401.
regime, water availability and land use/land cover change impact on the water Thomaz, E. L., Vestena, L. R., & Ramos Scharro n, C. E. (2014). The effects of unpaved
balance in a large agriculture basin in the Southern Brazilian Amazon. Journal of roads on suspended sediment concentration at varying spatial scalesea case
South American Earth Sciences, 108, Article 103224. study from South Brazil. Water and Environment Journal, 28(4), 547e555.
Lo pez-Ballesteros, A., Senent-Aparicio, J., Srinivasan, R., & Pe rez-Sa
nchez, J. (2019). Tiecher, T., Caner, L., Minella, J. P. G., & dos Santos, D. R. (2015). Combining visible-
Assessing the impact of best management practices in a highly anthropogenic based-color parameters and geochemical tracers to improve sediment source
and ungauged watershed using the SWAT model: A case study in the el beal discrimination and apportionment. Science of the Total Environment, 527,
watershed (Southeast Spain). Agronomy, 9(10), 576. 135e149.
Mallet, F., Marc, V., Douvinet, J., Rossello, P., Joly, D., & Ruy, S. (2020). Assessing soil Tiecher, T., Caner, L., Minella, J. P. G., Pellegrini, A., Capoane, V., Rasche, J. W. A., …
water content variation in a small mountainous catchment over different time dos Santos Rheinheimer, D. (2017). Tracing sediment sources in two paired
steps and land covers using geographical variables. Journal of Hydrology, 591, agricultural catchments with different riparian forest and wetland proportion
Article 125593. in South Brazil. Geoderma, 285, 225e239.
Mentges, M. I., Reichert, J. M., Rodrigues, M. F., Awe, G. O., & Mentges, L. R. (2016). Tiecher, T., Minella, J. P. G., Evrard, O., Caner, L., Merten, G. H., Capoane, V., … dos
Capacity and intensity soil aeration properties affected by granulometry, Santos, D. R. (2018). Fingerprinting sediment sources in a large agricultural
moisture, and structure in no-tillage soils. Geoderma, 263, 47e59. catchment under no-tillage in South Brazil (Conceiç~ ao River). Land Degradation
Minella, J. P. G., Merten, G. H., Reichert, J. M., & dos Santos, D. R. (2007). Identificaç~ ao & Development, 29(4), 939e951.
e implicaço ~es para a conservaça ~o do solo das fontes de sedimentos em bacias Tornquist, C. G. (2007). Simulaça ~o da dina^mica do carbono orga ^nico do solo em escala
hidrogra ficas. Revista Brasileira de Ci^ encia do Solo, 31, 1637e1646. regional: Aplicaça ~o do modelo Century e sistemas de informaço ~es geograficas
Minella, J. P. G., Merten, G. H., Walling, D. E., & Reichert, J. M. (2009). Changing (Doctoral dissertation). Porto Alegre, RS: Universidade Federal do Rio Grande do
sediment yield as an indicator of improved soil management practices in Sul.
southern Brazil. Catena, 79, 228e236. Tuppad, P., Kannan, N., Srinivasan, R., Rossi, C. G., & Arnold, J. G. (2010). Simulation
Morgan, R. P. C. (2009). Soil erosion and conservation. John Wiley & Sons. of agricultural management alternatives for watershed protection. Water Re-
Moriasi, D. N., Arnold, J. G., Van Liew, M. W., Bingner, R. L., Harmel, R. D., & sources Management, 24, 3115e3144.
15
Ebling et al.
Uniyal, B., Jha, M. K., Verma, A. K., & Anebagilu, P. K. (2020). Identification of critical Agriculture Handbook.
areas and evaluation of best management practices using SWAT for sustainable Wolschick, N. H., Bertol, I., Barbosa, F. T., Bagio, B., & Biasiolo, L. A. (2021). Remaining
watershed management. Science of the Total Environment, 744, Article 140737. effect of long-term soil tillage on plant biomass yield and water erosion in a
USDA, United States Department of Agriculture. (2017). Conservation practice stan- Cambisol after transition to no-tillage. Soil and Tillage Research, 213, Article
dard e contour farming. Retrieved from https://www.nrcs.usda.gov/wps/portal/ 105149.
nrcs/main/national/technical/cp/ncps/. Yang, X., Zheng, L., Yang, Q., Wang, Z., Cui, S., & Shen, Y. (2018). Modelling the effects
USDA-SCS (United States Department of AgricultureeSoil Conservation Service). of conservation tillage on crop water productivity, soil water dynamics and
(1972). National engineering handbook, section 4 hydrology. WashingMg, DC: evapotranspiration of a maize-winter wheat-soybean rotation system on the
USDA. Loess Plateau of China using APSIM. Agricultural Systems, 166, 111e123.
Vache, K. B., Eilers, J. M., & Santelmann, M. V. (2002). Water quality modeling of Yifru, B. A., Chung, I. M., Kim, M. G., & Chang, S. W. (2021). Assessing the effect of
alternative agricultural scenarios in the us corn belt 1. JAWRA Journal of the land/use land cover and climate change on water yield and groundwater
American Water Resources Association, 38(3), 773e787. recharge in east african rift valley using integrated model. Journal of Hydrology:
Vaz, C. M. P., de Freitas Iossi, M., de Mendonça Naime, J., Macedo, A., Reichert, J. M., Regional Studies, 37, Article 100926.
Reinert, D. J., & Cooper, M. (2005). Validation of the Arya and Paris water Zanon, J. A., Favaretto, N., Goularte, G. D., Dieckow, J., & Barth, G. (2020). Manure
retention model for Brazilian soils. Soil Science Society of America Journal, 69(3), application at long-term in no-till: Effects on runoff, sediment and nutrients
577e583. losses in high rainfall events. Agricultural Water Management, 228, Article
Waidler, D., White, M., Steglich, E., Wang, S., Williams, J., Jones, C. A., & Srinivasan, R. 105908.
(2011). Conservation practice modeling guide for SWAT and APEX. Texas Water Zhang, G. S., Chan, K. Y., Oates, A., Heenan, D. P., & Huang, G. B. (2007). Relationship
Resources Institute. between soil structure and runoff/soil loss after 24 years of conservation tillage.
Wang, J., Wu, T., Li, Q., & Wang, S. (2021). Quantifying the effect of environmental Soil and Tillage Research, 92(1e2), 122e128.
drivers on water conservation variation in the eastern Loess Plateau, China. Zhang, X., Song, J., Wang, Y., Deng, W., & Liu, Y. (2021). Effects of land use on slope
Ecological Indicators, 125, Article 107493. runoff and soil loss in the loess plateau of China: A meta-analysis. The Science of
Williams, J. R. (1995). Chapter 25: The EPIC model. In V. P. Singh (Ed.), Computer the Total Environment, 755, Article 142418.
models of watershed hydrology. Water Resources Publications. Zhao, L., Hou, R., & Wu, F. (2018). Effect of tillage on soil erosion before and after rill
Wischmeier, W., & Smith, D. (1978). Predicting rainfall erosion loss: A guide to con- development. Land Degradation & Development, 29(8), 2506e2513.
servation planning United States department of agriculture. WashingMg:
16

Thais Palumbo 2023

Uploaded by

Copyright:

Available Formats

Thais Palumbo 2023

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Thais Palumbo 2023

Uploaded by

Copyright:

Available Formats

International Soil and Water Conservation Research xxx (xxxx) xxx

Contents lists available at ScienceDirect

International Soil and Water Conservation Research

Original Research Article

Best management practices to reduce soil erosion and change water

1. Introduction South America, Southeast Asia, and Sub-Saharan Africa, where

Scenarios BMP type Combinations of BMP's

Base Scenario e Without BMP's

Calibrated variable Parameters Name Fitted value

Streamﬂow Baseﬂow alpha factor (days) v__ALPHA_BF.gw 0.09 0.26

Sediment yield Channel erodibility factor v__CH_COV1.rte 0.25 0.50

Reduction of SY varied from 20 to 60% in this scenario, in which 96% 4. Discussion

Studied paired watersheds consist predominantly of soil classes

You might also like