Nothing Special   »   [go: up one dir, main page]
Next Issue
Volume 8, July
Previous Issue
Volume 8, May
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
8.3
4.2
Journals
Remote Sensing
Volume 8
Issue 6
remotesensing-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Remote Sens., Volume 8, Issue 6 (June 2016) – 90 articles

Cover Story (view full-size image): As one of the most water-stressed cities in the world, Beijing has been suffering from land subsidence since 1935 due to over-exploitation of groundwater. Based on detailed analyses of Envisat ASAR images acquired between 2003 and 2010 and TerraSAR-X images from 2010 to 2011, this paper provides new insights into the spatio-temporal distribution characteristics and the key triggering and conditioning factors of land subsidence in Beijing. The maximum subsidence is observed in the eastern part of Beijing with a rate greater than 100 mm/year. The good agreement between InSAR and GPS results suggests that InSAR is a powerful tool for monitoring land subsidence. Interesting relationships in terms of land subsidence were found with groundwater level, active faults, accumulated soft soil thickness, aquifer types and the distances to pumping wells. View this paper
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
15942 KiB  
Article
Deformation and Related Slip Due to the 2011 Van Earthquake (Turkey) Sequence Imaged by SAR Data and Numerical Modeling
by Elisa Trasatti, Cristiano Tolomei, Giuseppe Pezzo, Simone Atzori and Stefano Salvi
Remote Sens. 2016, 8(6), 532; https://doi.org/10.3390/rs8060532 - 22 Jun 2016
Cited by 7 | Viewed by 7379
Abstract
A Mw 7.1 earthquake struck the Eastern Anatolia, near the city of Van (Turkey), on 23 October 2011. We investigated the coseismic surface displacements using the InSAR technique, exploiting adjacent ENVISAT tracks and COSMO-SkyMed images. Multi aperture interferometry was also applied, measuring ground [...] Read more.
A Mw 7.1 earthquake struck the Eastern Anatolia, near the city of Van (Turkey), on 23 October 2011. We investigated the coseismic surface displacements using the InSAR technique, exploiting adjacent ENVISAT tracks and COSMO-SkyMed images. Multi aperture interferometry was also applied, measuring ground displacements in the azimuth direction. We solved for the fault geometry and mechanism, and we inverted the slip distribution employing a numerical forward model that includes the available regional structural data. Results show a horizontally elongated high slip area (7–9 m) at 12–17 km depth, while the upper part of the fault results unruptured, enhancing its seismogenic potential. We also investigated the post-seismic phase acquiring most of the available COSMO-SkyMed, ENVISAT and TERRASAR-X SAR images. The computed afterslip distributions show that the shallow section of the fault underwent considerable aseismic slip during the early days after the mainshock, of tens of centimeters. Our results support the hypothesis of a seismogenic potential reduction within the first 8–10 km of the fault through the energy release during the post-seismic phase. Despite non-optimal data coverage and coherence issues, we demonstrate that useful information about the Van earthquake could still be retrieved from SAR data through detailed analysis. Full article
(This article belongs to the Special Issue Earth Observations for Geohazards)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Map of the Mw 7.1 Van earthquake (green star) and related aftershocks until the end of 2011. The Edremit-Van earthquake epicenter is indicated with the purple star. The aftershocks (green before the Edremit-Van earthquake, purple after) are from [<a href="#B2-remotesensing-08-00532" class="html-bibr">2</a>] and KOERI. The available focal mechanisms of M &gt; 5.0 events from Global-CMT earthquake catalog and [<a href="#B3-remotesensing-08-00532" class="html-bibr">3</a>] are shown. Near field GPS stations (MURA and OZAL) are indicated by black triangles.</p> Full article ">Figure 2
<p>Graphical representation of the interferograms’ time span during the coseismic and post-seismic phases.</p> Full article ">Figure 3
<p>InSAR results: (<b>a</b>) CSK range results (<span class="html-italic">i.e.</span>, LOS direction); CSK1; (<b>b</b>) CSK azimuth results (positive Northwards); (<b>c</b>) ENV1, LOS displacements; and (<b>d</b>) ENV2, LOS displacements. SAR data details can be found in <a href="#remotesensing-08-00532-t001" class="html-table">Table 1</a>; (<b>e</b>) Data along the profile AA’: CSK1 (range measurements, black), CSK (azimuth measurements, red), and ENV2 (LOS measurements, blue). The grey vertical band indicates the fault trace (F1) and its uncertainty (~1 km). The green star is the hypocenter.</p> Full article ">Figure 4
<p>Fractures and landslides from InSAR: (<b>A</b>) wrapped phase of the CSK interferogram, where some crack lineaments (FA and FB, purple lines) and landslides (blue lines) are recognized; (<b>B</b>) zoomed view of the landslides; (<b>C</b>) Google Earth view from South of landslide 1 (red lines); and in green the displacement profile shown in (<b>D</b>), where the two red bars indicate the DGSD body; (<b>E</b>) Google Earth view from NNW of landslides 3 and 4, indicated in red.</p> Full article ">Figure 5
<p>FE model of the Van Earthquake: (<b>a</b>) Top view of the model. The uniform slip fault is reported by green line while the fault plane used to retrieve the slip distribution is indicated by the black line. Surface ruptures constrained by InSAR are indicated by purple lines. The green star is the hypocenter and the orange dots are the surface nodes of the FE model; (<b>b</b>) Rigidity distribution on the section AA’ (see <a href="#remotesensing-08-00532-f003" class="html-fig">Figure 3</a>) and on the fault plane within the heterogeneous FE model, view from West. The external edges of the FE model are shown with ochre lines. The seismicity within few kilometers from the section is reported by ochre spheres.</p> Full article ">Figure 6
<p>Fault slip distribution retrieved from geodetic data: (<b>a</b>) fault slip distribution; and (<b>b</b>) error associated. The surface fractures are reported by purple lines, and the Van earthquake epicenter by the green star.</p> Full article ">Figure 7
<p>Comparisons between observed data (first column) and predictions (second column) related to the coseismic displacements of the Van earthquake. The residuals are observed minus modeled data (third column): (<b>a</b>–<b>c</b>) CSK1 (1210 data points); (<b>d</b>–<b>f</b>) CSK azimuth data (891 points); (<b>g</b>–<b>i</b>) ENV1 (2317 data points), near field GPS displacement vectors are also reported (black, observed; red, computed); and (<b>j</b>–<b>l</b>) ENV2 (896 data points). The green star is the hypocenter while the fault embedded in the FE model is indicated in black, with slip contour each 2 m.</p> Full article ">Figure 7 Cont.
<p>Comparisons between observed data (first column) and predictions (second column) related to the coseismic displacements of the Van earthquake. The residuals are observed minus modeled data (third column): (<b>a</b>–<b>c</b>) CSK1 (1210 data points); (<b>d</b>–<b>f</b>) CSK azimuth data (891 points); (<b>g</b>–<b>i</b>) ENV1 (2317 data points), near field GPS displacement vectors are also reported (black, observed; red, computed); and (<b>j</b>–<b>l</b>) ENV2 (896 data points). The green star is the hypocenter while the fault embedded in the FE model is indicated in black, with slip contour each 2 m.</p> Full article ">Figure 8
<p>Post-seismic InSAR results from CSK and TSX satellites: (<b>a</b>) CSK2; (<b>b</b>) CSK3; (<b>c</b>) TSX1; (<b>d</b>) TSX2; (<b>e</b>) TSX3; and (<b>f</b>) difference between CSK3 and TSX2 (see text and <a href="#remotesensing-08-00532-t001" class="html-table">Table 1</a> for details). Master/slave dates are indicated, along with the days after the mainshock in brackets. The green star is the mainshock epicenter and the purple star is the Edremit-Van earthquake epicenter (shown only if included in the temporal baseline).</p> Full article ">Figure 9
<p>Post-seismic InSAR data and modeling (three-day temporal baseline): (<b>a</b>) subsampled CSK2 data (step of 500 m, 5660 data points); (<b>b</b>) modeled LOS displacements and the fault afterslip contour each 10 cm; (<b>c</b>) residuals (observed minus modeled data); and (<b>d</b>) related afterslip distribution. The green star is the mainshock epicenter.</p> Full article ">Figure 10
<p>Post-seismic InSAR data and modeling (17-days temporal baseline): (<b>a</b>) subsampled data (step of 500 m, 1590 data points) computed from the difference between images CSK3 and TSX2 (see text for details); (<b>b</b>) modeled LOS displacements and the fault afterslip contour each 10 cm; (<b>c</b>) residuals (observed minus modeled data); and (<b>d</b>) related afterslip distribution. The green star is the mainshock epicenter.</p> Full article ">Figure 11
<p>Coseismic and post-seismic results: (<b>a</b>) 3D view of the Van earthquake fault from the NW. Coseismic slip (white lines, contour every 2 m) and the post-seismic slip (red and black lines are three and 17 days after the mainshock, respectively, contour every 10 cm) superimposed on the rigidity heterogeneities of the fault plane within the FE model. Aftershocks hypocenters within 4 km from the fault are shown by ochre spheres; (<b>b</b>) AA’ profile (see <a href="#remotesensing-08-00532-f005" class="html-fig">Figure 5</a>a) across the traces of the main fault (F1) and the splay fault (F2). The colors are: CSK2, black; CSK3, blue; TSX1, grey; TSX2, yellow; TSX3, green; and the difference between CSK3 and TSX2, red. The green star is the Van earthquake epicenter.</p> Full article ">
7163 KiB  
Article
Quantifying Fertilizer Application Response Variability with VHR Satellite NDVI Time Series in a Rainfed Smallholder Cropping System of Mali
by Xavier Blaes, Guillaume Chomé, Marie-Julie Lambert, Pierre Sibiry Traoré, Antonius G. T. Schut and Pierre Defourny
Remote Sens. 2016, 8(6), 531; https://doi.org/10.3390/rs8060531 - 22 Jun 2016
Cited by 21 | Viewed by 11247
Abstract
Soil fertility in smallholder farming areas is known to vary strongly on multiple scales. This study measures the sensitivity of the recorded satellite signal to on-farm soil fertility treatments applied to five crop types, and quantifies this fertilization effect with respect to within-field [...] Read more.
Soil fertility in smallholder farming areas is known to vary strongly on multiple scales. This study measures the sensitivity of the recorded satellite signal to on-farm soil fertility treatments applied to five crop types, and quantifies this fertilization effect with respect to within-field variation, between-field variation and field position in the catena. Plant growth was assessed in 5–6 plots per field in 48 fields located in the Sudano-Sahelian agro-ecological zone of southeastern Mali. A unique series of Very High Resolution (VHR) satellite and Unmanned Aerial Vehicle (UAV) images were used to calculate the Normalized Difference Vegetation Index (NDVI). In this experiment, for half of the fields at least 50% of the NDVI variance within a field was due to fertilization. Moreover, the sensitivity of NDVI to fertilizer application was crop-dependent and varied through the season, with optima at the end of August for peanut and cotton and early October for sorghum and maize. The influence of fertilizer on NDVI was comparatively small at the landscape scale (up to 35% of total variation), relative to the influence of other components of variation such as field management and catena position. The NDVI response could only partially be benchmarked against a fertilization reference within the field. We conclude that comparisons of the spatial and temporal responses of NDVI, with respect to fertilization and crop management, requires a stratification of soil catena-related crop growth conditions at the landscape scale. Full article
(This article belongs to the Special Issue Remote Sensing in Precision Agriculture)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Sukumba site, near Koutiala in southeastern Mali. The 10 km × 10 km blue box locates the VHR imagery acquisition. The <span class="html-italic">in-situ</span> bi-weekly monitored fields represented by green boundaries are spread along a catena transect from plateau (north) to lowland (south). Background imagery is a GeoEye RGB color composite from 24 June 2014.</p> Full article ">Figure 2
<p>Schematic layout of the experiment in a hypothetical field of 1 ha.</p> Full article ">Figure 3
<p>Mean and standard deviation (colored shading) of NDVI recorded in plots B–F, reflecting different fertilizer application treatments, for different overpass dates.</p> Full article ">Figure 4
<p>False color composites (NIR band seen in red color, red band in green and green band in blue) showing the plot effect on canopy development: late August GeoEye image at 2 m resolution (<b>a</b>,<b>b</b>); late August UAV image at 0.1 m resolution (<b>c</b>,<b>d</b>); a millet field with a sandy soil (<b>a</b>–<b>c</b>); and a sorghum field with a sandy-clay soils (<b>b</b>–<b>d</b>).</p> Full article ">Figure 5
<p>Evolution of crop height (<b>a</b>,<b>b</b>) and ground coverage (GC) fraction (<b>c</b>,<b>d</b>) for fertilizer application treatments A–F for the Millet (<b>a</b>–<b>c</b>) and Sorghum (<b>b</b>–<b>d</b>) fields also shown in <a href="#remotesensing-08-00531-f004" class="html-fig">Figure 4</a>.</p> Full article ">Figure 6
<p>R<sup>2</sup> boxplot from the model describing the effect of the different fertilization levels on NDVI. The grey vertical strips indicate the fertilizers application window for each crop. The red boxplots correspond to the date where crop reaction to fertilization treatments is strongest. The black dots represent extreme R<sup>2</sup> values that were identified as outliers.</p> Full article ">Figure 7
<p>Crop specific semi-variograms fitted on NDVI values extracted from the 10 September 2014 UAV imagery with about 10 cm resolution.</p> Full article ">Figure 8
<p>Proportion of the fields with a significant difference in mean NDVI between the fertility treatment pairs at the time of the largest observed fertilization responses: 26 August 2014 (peanut, millet and sorghum) and 4 October 2014 (maize and cotton).</p> Full article ">Figure 9
<p>Relationships between plot mean NDVI extracted from the DG image acquired on 26 August 2014 and the green ground coverage (GC) as measured in the field around this date (±3 days) for the five crop types.</p> Full article ">Figure 10
<p>Relationships between plot mean NDVI extracted from the DG image of 26 August 2014 and mean plant height within plots as measured in the field around this date (±3 days) for the five crop types.</p> Full article ">Figure 11
<p>Proportion of the total NDVI variance explained by the three components of spatial variation (stratum, field and plot) for five crops. The sowing and fertilization application windows are shown for each crop as green and grey shade, respectively.</p> Full article ">
8459 KiB  
Article
Sixteen Years of Agricultural Drought Assessment of the BioBío Region in Chile Using a 250 m Resolution Vegetation Condition Index (VCI)
by Francisco Zambrano, Mario Lillo-Saavedra, Koen Verbist and Octavio Lagos
Remote Sens. 2016, 8(6), 530; https://doi.org/10.3390/rs8060530 - 22 Jun 2016
Cited by 78 | Viewed by 9879
Abstract
Drought is one of the most complex natural hazards because of its slow onset and long-term impact; it has the potential to negatively affect many people. There are several advantages to using remote sensing to monitor drought, especially in developing countries with limited [...] Read more.
Drought is one of the most complex natural hazards because of its slow onset and long-term impact; it has the potential to negatively affect many people. There are several advantages to using remote sensing to monitor drought, especially in developing countries with limited historical meteorological records and a low weather station density. In the present study, we assessed agricultural drought in the croplands of the BioBío Region in Chile. The vegetation condition index (VCI) allows identifying the temporal and spatial variations of vegetation conditions associated with stress because of rainfall deficit. The VCI was derived at a 250 m spatial resolution for the 2000–2015 period with the Moderate Resolution Imaging Spectroradiometer (MODIS) MOD13Q1 product. We evaluated VCI for cropland areas using the land cover MCD12Q1 version 5.1 product and compared it to the in situ Standardized Precipitation Index (SPI) for six-time scales (1–6 months) from 26 weather stations. Results showed that the 3-month SPI (SPI-3), calculated for the modified growing season (November–April) instead of the regular growing season (September–April), has the best Pearson correlation with VCI values with an overall correlation of 0.63 and between 0.40 and 0.78 for the administrative units. These results show a very short-term vegetation response to rainfall deficit in September, which is reflected in the vegetation in November, and also explains to a large degree the variation in vegetation stress. It is shown that for the last 16 years in the BioBío Region we could identify the 2007/2008, 2008/2009, and 2014/2015 seasons as the three most important drought events; this is reflected in both the overall regional and administrative unit analyses. These results concur with drought emergencies declared by the regional government. Future studies are needed to associate the remote sensing values observed at high resolution (250 m) with the measured crop yield to identify more detailed individual crop responses. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Study area. (<b>a</b>) BioBío Region administrative units in a digital terrain model with 26 weather stations; (<b>b</b>) Location of the BioBío Region, Chile.</p> Full article ">Figure 2
<p>Bioclimatic precipitation variables of the BioBío Region, Chile. (<b>a</b>) Direst month; (<b>b</b>) Wettest month; (<b>c</b>) Annual preciaitation.</p> Full article ">Figure 3
<p>Bioclimatic temperature variables of the BioBío Region, Chile. (<b>a</b>) Direst month; (<b>b</b>) Wettest month; (<b>c</b>) Annual preciaitation.</p> Full article ">Figure 4
<p>Land cover classes in the BioBío Region, Chile, based on the IGBP land cover scheme for the MODIS MCD12Q1 version 5.1 product.</p> Full article ">Figure 5
<p>Time-series comparison between raw NDVI (points) and smoothed NDVI by Lowess (lines) for five points on five administrative units in cropland areas of the BioBío Region.Time-series comparison between raw NDVI (points) and smoothed NDVI by Lowess (lines) for five points on five administrative units in cropland areas of the BioBío Region</p> Full article ">Figure 6
<p>Variation of the (<b>a</b>) global VCI percentage (%) of cropland surface with different VCI classes and (<b>b</b>) boxplot of global VCI intensity (%) for the growing seasons between 2000/2001 and 2014/2015 in the Biobío Region, Chile.Variation of the (a) global VCI percentage (%) of cropland surface with different VCI classes and (b) boxplot of global VCI intensity (%) for the growing seasons between 2000/2001 and 2014/2015 in the Biobío Region, Chile</p> Full article ">Figure 7
<p>Heatmap of VCI intensity and percentage of cropland area &lt; VCI = 40 in the main cropland administrative units in the study area. (<b>a</b>) Intensity and (<b>b</b>) Surface under drought for the 2014–2015 season; (<b>c</b>) Intensity and (<b>d</b>) Surface under drought for 2007–2009 season. The dashed white line corresponds to the date when the agricultural drought emergency was declared by decision makers.</p> Full article ">Figure 8
<p>(<b>a</b>) Mean VCI conditions and (<b>b</b>) percentage of cropland surface with VCI ≤ 40% in the administrative units of the Biobío Region, Chile, for the 2000/2001, 2007/2008, 2008/2009, and 2014/2015 growing seasons (September–April).</p> Full article ">Figure 9
<p>VCI mean values for croplands during growing season (September–April) in the Biobío Region, Chile, for 2007/2008, 2008/2009, and 2014/2015 seasons.</p> Full article ">Figure 10
<p>Comparison of SPI-3 and VCI anomaly for 15 administrative units with percentage cropland &gt; 10% from 2000/2001 to 2014/2015 modified growing seasons (November–April).</p> Full article ">Figure 11
<p>(<b>a</b>) Correlation between mean VCI (croplands) and mean SPI at time scales between 1 to 6 for meteorological station Number 4 in the Biobío Region, Chile, for three different periods; (<b>b</b>) Monthly correlation between SPI-1, SPI-3 and SPI-6 with VCI in the growing season. Letter (<b>a</b>) in the plot means significance at <span class="html-italic">p</span> &gt; 0.01.</p> Full article ">
2872 KiB  
Article
Posterior Probability Modeling and Image Classification for Archaeological Site Prospection: Building a Survey Efficacy Model for Identifying Neolithic Felsite Workshops in the Shetland Islands
by William P. Megarry, Gabriel Cooney, Douglas C. Comer and Carey E. Priebe
Remote Sens. 2016, 8(6), 529; https://doi.org/10.3390/rs8060529 - 22 Jun 2016
Cited by 14 | Viewed by 6350
Abstract
The application of custom classification techniques and posterior probability modeling (PPM) using Worldview-2 multispectral imagery to archaeological field survey is presented in this paper. Research is focused on the identification of Neolithic felsite stone tool workshops in the North Mavine region of the [...] Read more.
The application of custom classification techniques and posterior probability modeling (PPM) using Worldview-2 multispectral imagery to archaeological field survey is presented in this paper. Research is focused on the identification of Neolithic felsite stone tool workshops in the North Mavine region of the Shetland Islands in Northern Scotland. Sample data from known workshops surveyed using differential GPS are used alongside known non-sites to train a linear discriminant analysis (LDA) classifier based on a combination of datasets including Worldview-2 bands, band difference ratios (BDR) and topographical derivatives. Principal components analysis is further used to test and reduce dimensionality caused by redundant datasets. Probability models were generated by LDA using principal components and tested with sites identified through geological field survey. Testing shows the prospective ability of this technique and significance between 0.05 and 0.01, and gain statistics between 0.90 and 0.94, higher than those obtained using maximum likelihood and random forest classifiers. Results suggest that this approach is best suited to relatively homogenous site types, and performs better with correlated data sources. Finally, by combining posterior probability models and least-cost analysis, a survey least-cost efficacy model is generated showing the utility of such approaches to archaeological field survey. Full article
(This article belongs to the Special Issue Archaeological Prospecting and Remote Sensing)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Northwestern Europe, the Shetland Islands and the North Roe Felsite Project (NRFP) Extent (Images by the Authors).</p> Full article ">Figure 2
<p>A Felsite Workshop at the Beorgs of Uyea. Felsite is visible in blue against Red Ronas Granite (Images by the Authors).</p> Full article ">Figure 3
<p>Graphical posterior probability model (PPM) generation steps.</p> Full article ">Figure 4
<p>Pair Plots for Model 6 components 1, 2 and 3.</p> Full article ">Figure 5
<p>Sample receiver operating characteristic (ROC) curve for Model 7 (with area under the curve (AUC) 0.9611.</p> Full article ">Figure 6
<p>Probability Model 9 and Hybrid surface using Cost-Surface.</p> Full article ">Figure 7
<p>The effects of principal components analysis on 8 Worldview-2 bands, showing workshops at the Beorgs of Uyea.</p> Full article ">
1704 KiB  
Article
An IPCC-Compliant Technique for Forest Carbon Stock Assessment Using Airborne LiDAR-Derived Tree Metrics and Competition Index
by Chinsu Lin, Gavin Thomson and Sorin C. Popescu
Remote Sens. 2016, 8(6), 528; https://doi.org/10.3390/rs8060528 - 22 Jun 2016
Cited by 48 | Viewed by 7071
Abstract
This study developed an IPCC (Intergovernmental Panel on Climate Change) compliant method for the estimation of above-ground carbon (AGC) in forest stands using remote sensing technology. A multi-level morphological active contour (MMAC) algorithm was employed to obtain tree-level metrics (tree height (LH), crown [...] Read more.
This study developed an IPCC (Intergovernmental Panel on Climate Change) compliant method for the estimation of above-ground carbon (AGC) in forest stands using remote sensing technology. A multi-level morphological active contour (MMAC) algorithm was employed to obtain tree-level metrics (tree height (LH), crown radius (LCR), competition index (LCI), and stem diameter (LDBH)) from an airborne LiDAR-derived canopy height model. Seven biomass-based AGC models and 13 volume-based AGC models were developed using a training dataset and validated using a separate validation dataset. Four accuracy measures, mean absolute error (MAE), root-mean-square error (RMSE), percentage RMSE (PRMSE), and root-mean-square percentage error (RMSPE) were calculated for each of the 20 models. These measures were transformed into a new index, accuracy improvement percentage (AIP), for post hoc testing of model performance in estimating forest stand AGC stock. Results showed that the tree-level AGC models explained 84% to 91% of the variance in tree-level AGC within the training dataset. Prediction errors (RMSEs) for these models ranged between 15 ton/ha and 210 ton/ha in mature forest stands, which is equal to an error percentage in the range 6% to 86%. At the stand-level, several models achieved accurate and reliable predictions of AGC stock. Some models achieved 90% to 95% accuracy, which was equal to or superior to the R-squared of the tree-level AGC models. The first recommended model was a biomass-based model using the metrics LDBH, LH, and LCI and the others were volume-based models using LH, LCI, and LCR and LDBH and LH. One metric, LCI, played a critical role in upgrading model performance when banded together with LH and LCR or LDBH and LCR. We conclude by proposing an IPCC-compatible method that is suitable for calculating tree-level AGC and predicting AGC stock of forest stands from airborne LiDAR data. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Illustration of the study site location (<b>a</b>) and its ortho-rectified bitmap image (<b>b</b>) and canopy height model (<b>c</b>).</p> Full article ">Figure 2
<p>Flowchart of the above-ground-biomass carbon accounting process.</p> Full article ">Figure 3
<p>The relationship between a subject tree and its competitors. (<b>a</b>) The proximity area is defined based on the location of the subject tree and a radius <b><span class="html-italic">r</span></b>; (<b>b</b>) The angle <math display="inline"> <semantics> <mrow> <msub> <mi>θ</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> </mrow> </semantics> </math> is the competition pressure exerted on a subject tree <b><span class="html-italic">i</span></b> by a neighboring tree <b><span class="html-italic">j</span></b> in the proximity area.</p> Full article ">Figure 4
<p>Prediction error variations of the tree-level AGC models in predicting stand-level AGC stocks.</p> Full article ">
1048 KiB  
Erratum
Erratum: Dupuy, E.; et al. Comparison of XH2O Retrieved from GOSAT Short-Wavelength Infrared Spectra with Observations from the TCCON Network. Remote Sensing 2016, 8, 414
by Remote Sensing Editorial Office
Remote Sens. 2016, 8(6), 527; https://doi.org/10.3390/rs8060527 - 22 Jun 2016
Viewed by 3265
Abstract
In the published paper [1], the plot sizes of Figures 4, 6 and 9 were incorrect.[...] Full article
►▼ Show Figures

Figure 4

Figure 4
<p>Scatter plot of the GOSAT TANSO-FTS XH<sub>2</sub>O and coincident TCCON soundings (criteria of ±30 min and ±1° in latitude/longitude). For these criteria, there are no coincident TANSO-FTS ocean scans. The caption and color-coding are identical to those of Figure 3.</p> Full article ">Figure 6
<p>Scatter plot of the GOSAT TANSO-FTS XH<sub>2</sub>O and coincident TCCON soundings (criteria of ±30 min and ±1° in latitude/longitude) at the Lamont (<b>left</b>) and Lauder (<b>right</b>) TCCON sites. All coincidences were found for TANSO-FTS land scans.</p> Full article ">Figure 9
<p>Relative differences (GOSAT-TCCON)/TCCON as a function of the difference, in meters, between the retrieved altitude of the GOSAT footprints and the altitude of the TCCON sites, for GOSAT land scans only. The caption and color-coding are identical to those of Figure 3.</p> Full article ">
923 KiB  
Article
Using Different Regression Methods to Estimate Leaf Nitrogen Content in Rice by Fusing Hyperspectral LiDAR Data and Laser-Induced Chlorophyll Fluorescence Data
by Lin Du, Shuo Shi, Jian Yang, Jia Sun and Wei Gong
Remote Sens. 2016, 8(6), 526; https://doi.org/10.3390/rs8060526 - 22 Jun 2016
Cited by 32 | Viewed by 6070
Abstract
Nitrogen is an essential nutrient element in crop photosynthesis and yield improvement. Thus, it is urgent and important to accurately estimate the leaf nitrogen contents (LNC) of crops for precision nitrogen management. Based on the correlation between LNC and reflectance spectra, the hyperspectral [...] Read more.
Nitrogen is an essential nutrient element in crop photosynthesis and yield improvement. Thus, it is urgent and important to accurately estimate the leaf nitrogen contents (LNC) of crops for precision nitrogen management. Based on the correlation between LNC and reflectance spectra, the hyperspectral LiDAR (HSL) system can determine three-dimensional structural parameters and biochemical changes of crops. Thereby, HSL technology has been widely used to monitor the LNC of crops at leaf and canopy levels. In addition, the laser-induced fluorescence (LIF) of chlorophyll, related to the histological structure and physiological conditions of green plants, can also be utilized to detect nutrient stress in crops. In this study, four regression algorithms, support vector machines (SVMs), partial least squares (PLS) and two artificial neural networks (ANNs), back propagation NNs (BP-NNs) and radial basic function NNs (RBF-NNs), were selected to estimate rice LNC in booting and heading stages based on reflectance and LIF spectra. These four regression algorithms were used for 36 input variables, including the reflectance spectral variables on 32 wavelengths and four peaks of the LIF spectra. A feature weight algorithm was proposed to select different band combinations for the LNC retrieval models. The determination coefficient (R2) and the root mean square error (RMSE) of the retrieval models were utilized to compare their abilities of estimating the rice LNC. The experimental results demonstrate that (I) these four regression methods are useful for estimating rice LNC in the order of RBF-NNs > SVMs > BP-NNs > PLS; (II) The LIF data in two forms, including peaks and indices, display potential in rice LNC retrieval, especially when using the PLS regression (PLSR) model for the relationship of rice LNC with spectral variables. The feature weighting algorithm is an effective and necessary method to determine appropriate band combinations for rice LNC estimation. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Example of a fluorescence spectrum and its approximation by a Gaussian model.</p> Full article ">Figure 2
<p><span class="html-italic">R</span><sup>2</sup> of estimation models and the 1:1 relationship between the observed LNC and those estimated using SVMs, ANNs (RBF-NN and BP-NN) and PLS in the (<b>a</b>) booting and (<b>b</b>) heading stages based on the reflectance spectra.</p> Full article ">Figure 3
<p>Different band combinations with an <span class="html-italic">R</span><sup>2</sup> of &gt;0.7 used for LNC estimation based on different regression methods in the (<b>a</b>) booting and (<b>b</b>) heading stages.</p> Full article ">
6604 KiB  
Article
Crop Monitoring Based on SPOT-5 Take-5 and Sentinel-1A Data for the Estimation of Crop Water Requirements
by Ana Navarro, João Rolim, Irina Miguel, João Catalão, Joel Silva, Marco Painho and Zoltán Vekerdy
Remote Sens. 2016, 8(6), 525; https://doi.org/10.3390/rs8060525 - 22 Jun 2016
Cited by 71 | Viewed by 10152
Abstract
Optical and microwave images have been combined for land cover monitoring in different agriculture scenarios, providing useful information on qualitative and quantitative land cover changes. This study aims to assess the complementarity and interoperability of optical (SPOT-5 Take-5) and synthetic aperture radar (SAR) [...] Read more.
Optical and microwave images have been combined for land cover monitoring in different agriculture scenarios, providing useful information on qualitative and quantitative land cover changes. This study aims to assess the complementarity and interoperability of optical (SPOT-5 Take-5) and synthetic aperture radar (SAR) (Sentinel-1A) data for crop parameter (basal crop coefficient (Kcb) values and the length of the crop’s development stages) retrieval and crop type classification, with a focus on crop water requirements, for an irrigation perimeter in Angola. SPOT-5 Take-5 images are used as a proxy of Sentinel-2 data to evaluate the potential of their enhanced temporal resolution for agricultural applications. In situ data are also used to complement the Earth Observation (EO) data. The Normalized Difference Vegetation Index (NDVI) and dual (VV + VH) polarization backscattering time series are used to compute the Kcb curve for four crop types (maize, soybean, bean and pasture) and to estimate the length of each phenological growth stage. The Kcb values are then used to compute the crop’s evapotranspiration and to subsequently estimate the crop irrigation requirements based on a soil water balance model. A significant R2 correlation between NDVI and backscatter time series was observed for all crops, demonstrating that optical data can be replaced by microwave data in the presence of cloud cover. However, it was not possible to properly identify each stage of the crop cycle due to the lack of EO data for the complete growing season. Full article
(This article belongs to the Special Issue Land Surface Processes and Interactions—From HCMM to Sentinel Missions and beyond)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>(<b>a</b>) Regional context of the test area (solid blue rectangle) with the location of the Kibala and Wako-Kungo weather stations of the Southern African Science Service Centre for Climate Change and Adaptive Land Management (SASSCAL) WeatherNet; (<b>b</b>) test area with the locations of the ground truth areas (yellow polygons) collected in April 2015.</p> Full article ">Figure 2
<p>Sentinel-1A and SPOT-5 Take-5 acquisition timeline, from 26 March 2015 (DOY 85) to 4 October 2015 (DOY 277).</p> Full article ">Figure 3
<p>Crop calendar of the main crops for the Wako-Kungo irrigation perimeter.</p> Full article ">Figure 4
<p>Flowchart of the IrrigRotation soil water balance model [<a href="#B22-remotesensing-08-00525" class="html-bibr">22</a>].</p> Full article ">Figure 5
<p>Basal crop coefficient (<span class="html-italic">K<sub>cb</sub></span>) curve (adapted from [<a href="#B13-remotesensing-08-00525" class="html-bibr">13</a>]).</p> Full article ">Figure 6
<p>Mean NDVI (<b>a</b>) and gamma VV (<b>b</b>) and VH (<b>c</b>) backscattering time series for maize, soybean, pasture and bean.</p> Full article ">Figure 7
<p>Scatterplots of the linear regression between the gamma VV + VH Sentinel-1A bands and the NDVI band, (<b>a</b>,<b>b</b>), respectively.</p> Full article ">Figure 8
<p>Overall accuracies and kappa coefficients (%) obtained from the cumulative addition of SPOT-5 bands into the NN classification process.</p> Full article ">Figure 9
<p>Individual class producer’s (PA) and user’s (UA) accuracies (%) for the best result from a multitemporal classification of the SPOT-5 bands using the neural network classifier.</p> Full article ">Figure 10
<p>Extract of a map corresponding to the best result from a multitemporal classification of the SPOT-5 bands using the neural network classifier.</p> Full article ">Figure 11
<p>Reference evapotranspiration (<span class="html-italic">ET<sub>o</sub></span>) and crop evapotranspiration (<span class="html-italic">ET<sub>c</sub></span>) for each crop (mm).</p> Full article ">Figure 12
<p>Total crop water requirement (m<sup>3</sup>·day<sup>−1</sup>) volume.</p> Full article ">
5978 KiB  
Article
Sea and Freshwater Ice Concentration from VIIRS on Suomi NPP and the Future JPSS Satellites
by Yinghui Liu, Jeffrey Key and Robert Mahoney
Remote Sens. 2016, 8(6), 523; https://doi.org/10.3390/rs8060523 - 22 Jun 2016
Cited by 43 | Viewed by 9746
Abstract
Information on ice is important for shipping, weather forecasting, and climate monitoring. Historically, ice cover has been detected and ice concentration has been measured using relatively low-resolution space-based passive microwave data. This study presents an algorithm to detect ice and estimate ice concentration [...] Read more.
Information on ice is important for shipping, weather forecasting, and climate monitoring. Historically, ice cover has been detected and ice concentration has been measured using relatively low-resolution space-based passive microwave data. This study presents an algorithm to detect ice and estimate ice concentration in clear-sky areas over the ocean and inland lakes and rivers using high-resolution data from the Visible Infrared Imaging Radiometer Suite (VIIRS) onboard the Suomi National Polar Orbiting Partnership (S-NPP) and on future Joint Polar Satellite System (JPSS) satellites, providing spatial detail that cannot be obtained with passive microwave data. A threshold method is employed with visible and infrared observations to identify ice, then a tie-point algorithm is used to determine the representative reflectance/temperature of pure ice, estimate the ice concentration, and refine the ice cover mask. The VIIRS ice concentration is validated using observations from Landsat 8. Results show that VIIRS has an overall bias of −0.3% compared to Landsat 8 ice concentration, with a precision (uncertainty) of 9.5%. Biases and precision values for different ice concentration subranges from 0% to 100% can be larger. Full article
(This article belongs to the Special Issue Sea Ice Remote Sensing and Analysis)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>High level processing flow diagram of the ice cover mask and concentration algorithm.</p> Full article ">Figure 2
<p>(<b>left</b>) Natural-color RGB image combination of Landsat 8 Operational Land Imager (OLI) band 5 (1.6 μm, red component), band 4 (0.86 μm, green component) and band 3 (0.64 μm, blue component) resampled to 50 m resolution; and (<b>right</b>) VIIRS M-band natural-color RGB image combination of band M10 (1.6 μm, red component), band M7 (0.86 μm, green component), and band M5 (0.67 μm, blue component) resampled to 1 km resolution. The images cover the Kara Sea east of Novaya Zemlya at 7:59 a.m. UTC on 25 June 2013. Sea ice and snow-covered sea ice are cyan; water is black; clouds are white.</p> Full article ">Figure 3
<p>Probability density distribution of 0.67 μm reflectance for the ice covered pixels shown in <a href="#remotesensing-08-00523-f002" class="html-fig">Figure 2</a>.</p> Full article ">Figure 4
<p>(<b>left</b>) Reflectance at 0.67 μm from S-NPP VIIRS; and (<b>right</b>) derived tie points at 0.67 μm.</p> Full article ">Figure 5
<p>Reflectance and sea ice concentration (SIC) relative frequency distribution at 0.64 μm for Landsat ice cover in <a href="#remotesensing-08-00523-f002" class="html-fig">Figure 2</a>.</p> Full article ">Figure 6
<p>(<b>left</b>) spatial distribution of the center point of 155 Landsat 8 scenes; and (<b>right</b>) the monthly distribution of scenes.</p> Full article ">Figure 7
<p>Ice concentration over the Arctic Ocean from the VIIRS overpass 8:43 p.m. to 9:03 p.m. UTC on 20 February 2015. The cloud-covered areas are white.</p> Full article ">Figure 8
<p>(<b>left</b>) Sea ice concentration derived from the Landsat image in <a href="#remotesensing-08-00523-f002" class="html-fig">Figure 2</a>; and (<b>right</b>) the calculated sea ice concentration using the Suomi NPP VIIRS data in <a href="#remotesensing-08-00523-f002" class="html-fig">Figure 2</a>. White areas denote pixels flagged out as either land or cloudy.</p> Full article ">Figure 9
<p>Ice concentration from SSMIS (<b>left</b>); and a daily ice concentration composite from VIIRS (<b>right</b>) over the Arctic on 20 February 2015. White areas in the SSMIS image denote pixels flagged as either land or the area around the pole that is not covered by the instrument. White areas in the VIIRS data denote pixels flagged as land, ice-free ocean, or cloud.</p> Full article ">Figure 10
<p>Ice concentration from SSMIS (<b>left</b>); and from VIIRS daily composite (<b>right</b>) over portion of the Arctic (longitude: 90–180, latitude: 70–90) on 20 February 2015. The North Pole is in the lower left corner. White areas in the SSMIS denote pixels flagged as either land or the area around the pole that is not covered by the instrument. White areas in the VIIRS data denote pixels flagged as either land or cloudy.</p> Full article ">Figure 11
<p>Aqua MODIS true-color image at 6:20 p.m. UTC on 28 March 2015 (<b>left</b>); and the corresponding ice concentration (<b>right</b>).</p> Full article ">Figure 12
<p>Comparison of VIIRS minus Landsat ice concentrations for different concentration ranges/bins. Note that for “All” and “90–100” plots the vertical axis range is different than for other sea ice concentration ranges.</p> Full article ">Figure 13
<p>The ice tie point adjustment scheme.</p> Full article ">Figure 14
<p>Comparison of VIIRS and Landsat ice concentrations for different concentration ranges/bins when a tie point adjustment scheme is employed.</p> Full article ">
2462 KiB  
Article
Sensitivity of L-Band SAR Backscatter to Aboveground Biomass of Global Forests
by Yifan Yu and Sassan Saatchi
Remote Sens. 2016, 8(6), 522; https://doi.org/10.3390/rs8060522 - 22 Jun 2016
Cited by 117 | Viewed by 12360
Abstract
Synthetic Aperture Radar (SAR) backscatter measurements are sensitive to forest aboveground biomass (AGB), and the observations from space can be used for mapping AGB globally. However, the radar sensitivity saturates at higher AGB values depending on the wavelength and geometry of radar measurements, [...] Read more.
Synthetic Aperture Radar (SAR) backscatter measurements are sensitive to forest aboveground biomass (AGB), and the observations from space can be used for mapping AGB globally. However, the radar sensitivity saturates at higher AGB values depending on the wavelength and geometry of radar measurements, and is influenced by the structure of the forest and environmental conditions. Here, we examine the sensitivity of SAR at the L-band frequency (~25 cm wavelength) to AGB in order to examine the performance of future joint National Aeronautics and Space Administration, Indian Space Research Organisation NASA-ISRO SAR mission in mapping the AGB of global forests. For SAR data, we use the Phased Array L-Band SAR (PALSAR) backscatter from the Advanced Land Observing Satellite (ALOS) aggregated at a 100-m spatial resolution; and for AGB data, we use more than three million AGB values derived from the Geoscience Laser Altimeter System (GLAS) LiDAR height metrics at about 0.16–0.25 ha footprints across eleven different forest types globally. The results from statistical analysis show that, over all eleven forest types, saturation level of L-band radar at HV polarization on average remains ≥100 Mg·ha−1. Fresh water swamp forests have the lowest saturation with AGB at ~80 Mg·ha−1, while needleleaf forests have the highest saturation at ~250 Mg·ha−1. Swamp forests show a strong backscatter from the vegetation-surface specular reflection due to inundation that requires to be treated separately from those on terra firme. Our results demonstrate that L-Band backscatter relations to AGB can be significantly different depending on forest types and environmental effects, requiring multiple algorithms to map AGB from time series of satellite radar observations globally. Full article
(This article belongs to the Special Issue Remote Sensing of Vegetation Structure and Dynamics)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Correlation between GLAS-derived AGB (Mg·ha<sup>−1</sup>) and ALOS backscatter coefficient sigma-0 for North American boreal forest. (<b>a</b>) shows all GLAS shots and corresponding sigma-0 values from the pixel that the GLAS shot falls on; (<b>b</b>) shows the mean (circle) and ±1 standard deviation (bar) of ALOS HV sigma-0 values by binning the shots into 5 Mg·ha<sup>−1</sup> bins. Bins below 10 Mg·ha<sup>−1</sup> and above 235 Mg·ha<sup>−1</sup> are removed as outliers.</p> Full article ">Figure 2
<p>Correlation between bin-averaged ALOS HV backscatter sigma-0 (m<sup>2</sup>·m<sup>−2</sup>) and middle of bin AGB value (Mg·ha<sup>−1</sup>) for (<b>a</b>) flooded forests; and (<b>b</b>) boreal forests. Outlier bins (such as some low bin values below 10 Mg·ha<sup>−1</sup>, and higher bins when the number of shots within the bin drop below 500, or 100 for the case of saline water flooded forests) are dropped. The flooded forests in (<b>a</b>) are divided into fresh-water and saline water (mangroves); boreal forests in (<b>b</b>) are divided by continent into North American boreal and Eurasian boreal.</p> Full article ">Figure 3
<p>Correlation between bin-averaged ALOS HV sigma-0 means and AGB values of the middle of the bin for the tropical moist forests of Africa, Southeast Asia, and Latin America, as well as temperature conifer forests. Solid lines represent the empirically fitted function of the form defined in Equation (2). Certain outliers were removed before fitting for the tropical moist forest (marked with x). Bins for tropical forests are cut off above 155 Mg·ha<sup>−1</sup> for the purpose of fitting.</p> Full article ">Figure 4
<p>Radar sensitivity required to detect a 20% change in AGB. The sensitivity requirement is calculated using Equations (3) and (4). The values for each bin are calculated using the derivatives from Equations (3) and (4) at the middle AGB value for each bin and ±10% from that middle AGB value. A total of 10 forest types are distinguished here, for better visibility, they are plotted in two separate figures: (<b>a</b>) Africa tropical moist broadleaf, South America tropical moist broadleaf, Asia tropical moist broadleaf, swamp forest/fresh water, mangrove/saline water (<b>b</b>) temperate conifer, Eurasia boreal, North America boreal, tropical savanna/shrub, and tropical dry broadleaf.</p> Full article ">Figure 5
<p>Radar sensitivity required to detect a 20 Mg·ha<sup>−1</sup> change in AGB. The sensitivity requirement is calculated using Equations (3) and (4). The values for each bin are calculated using the derivatives from Equations (3) and (4) at the middle AGB value for each bin and ±10 Mg·ha<sup>−1</sup> from that middle AGB value. A total of 10 forest types are distinguished here, for better visibility, they are plotted in two separate figures: (<b>a</b>) Africa tropical moist broadleaf, South America tropical moist broadleaf, Asia tropical moist broadleaf, swamp forest/fresh water, mangrove/saline water; (<b>b</b>) temperate conifer, Eurasia boreal, North America boreal, tropical savanna/shrub, and tropical dry broadleaf.</p> Full article ">Figure 6
<p>Number of radar observations required to observe a 20 Mg·ha<sup>−1</sup> change in AGB for various forest types. Values are calculated for each bin using the standard error in HV backscatter sigma-0 values to represent the amount of noise expected. For tropical savanna/shrubland (<b>b</b>), the number of looks is only calculated up to the AGB bin where enough data points are available. Forest types shown in (<b>a</b>) use a different y-axis scale as those shown in (<b>b</b>) due to the much lower observation requirements for the forests included in (<b>b</b>).</p> Full article ">Figure 7
<p>Average sigma-0 values (m<sup>2</sup>·m<sup>−2</sup>) for temperate conifer forests within 5 Mg·ha<sup>−1</sup> biomass bins. Red line is the fitted function using Equation (5) (with surface-volume scattering term) while green line is the fitted function using Equation (2) (with only canopy volume scattering term). The coefficients shown are those fitted for Equation (5) with ±1 standard deviation.</p> Full article ">Figure 8
<p>Statistical analysis for freshwater flooded forest category. Red markers are the bin-averaged ALOS HV sigma-0 values. Green histogram shows the total number of GLAS shots within each bin.</p> Full article ">
7479 KiB  
Article
Generation of Land Cover Maps through the Fusion of Aerial Images and Airborne LiDAR Data in Urban Areas
by Yongmin Kim
Remote Sens. 2016, 8(6), 521; https://doi.org/10.3390/rs8060521 - 22 Jun 2016
Cited by 7 | Viewed by 6251
Abstract
Satellite images and aerial images with high spatial resolution have improved visual interpretation capabilities. The use of high-resolution images has rapidly grown and has been extended to various fields, such as military surveillance, disaster monitoring, and cartography. However, many problems were encountered in [...] Read more.
Satellite images and aerial images with high spatial resolution have improved visual interpretation capabilities. The use of high-resolution images has rapidly grown and has been extended to various fields, such as military surveillance, disaster monitoring, and cartography. However, many problems were encountered in which one object has a variety of spectral properties and different objects have similar spectral characteristics in terms of land cover. The problems are quite noticeable, especially for building objects in urban environments. In the land cover classification process, these issues directly decrease the classification accuracy by causing misclassification of single objects as well as between objects. This study proposes a method of increasing the accuracy of land cover classification by addressing the problem of misclassifying building objects through the output-level fusion of aerial images and airborne Light Detection and Ranging (LiDAR) data. The new method consists of the following three steps: (1) generation of the segmented image via a process that performs adaptive dynamic range linear stretching and modified seeded region growth algorithms; (2) extraction of building information from airborne LiDAR data using a planar filter and binary supervised classification; and (3) generation of a land cover map using the output-level fusion of two results and object-based classification. The new method was tested at four experimental sites with the Min-Max method and the SSI-nDSM method followed by a visual assessment and a quantitative accuracy assessment through comparison with reference data. In the accuracy assessment, the new method exhibits various advantages, including reduced noise and more precise classification results. Additionally, the new method improved the overall accuracy by more than 5% over the comparative evaluation methods. The high and low patterns between the overall and building accuracies were similar. Thus, the new method is judged to have successfully solved the inaccuracy problem of classification that is often produced by high-resolution images of urban environments through an output-level fusion technique. Full article
(This article belongs to the Special Issue Multi-Sensor and Multi-Data Integration in Remote Sensing)
►▼ Show Figures

Figure 1

Figure 1
<p>Flowchart of classification generation using the new method.</p> Full article ">Figure 2
<p>Image segmentation: (<b>a</b>) original image; (<b>b</b>) radiometric-enhanced image; and (<b>c</b>) segmented image (red lines are segment boundaries).</p> Full article ">Figure 3
<p>Building information extracted from airborne LiDAR: (<b>a</b>) DTM; (<b>b</b>) building information (white areas).</p> Full article ">Figure 4
<p>Results of output-level fusion: (<b>a</b>) building area and (<b>b</b>) non-building area.</p> Full article ">Figure 5
<p>Aerial images of experimental sites: (<b>a</b>) site 1; (<b>b</b>) site 2; (<b>c</b>) site 3; and (<b>d</b>) site 4.</p> Full article ">Figure 6
<p>Final results for site 1: (<b>a</b>) aerial image; (<b>b</b>) Max-Min method; (<b>c</b>) SSI, nDSM method; and (<b>d</b>) the new method.</p> Full article ">Figure 7
<p>Results for site 2: (<b>a</b>) aerial image; (<b>b</b>) Max-Min method; (<b>c</b>) SSI, nDSM method; and (<b>d</b>) the new method.</p> Full article ">Figure 8
<p>Final results for site 3: (<b>a</b>) aerial image; (<b>b</b>) Max-Min method; (<b>c</b>) SSI, nDSM method; and (<b>d</b>) the new method.</p> Full article ">Figure 9
<p>Results for site 4: (<b>a</b>) aerial image; (<b>b</b>) Max-Min method; (<b>c</b>) SSI, nDSM method; and (<b>d</b>) the new method.</p> Full article ">Figure 10
<p>Overall accuracy of three classification results.</p> Full article ">Figure 11
<p>Producer’s and user’s accuracies for the building class: (<b>a</b>) site 1; (<b>b</b>) site 2; (<b>c</b>) site 3; and (<b>d</b>) site 4.</p> Full article ">
13973 KiB  
Article
Space Geodetic Observations and Modeling of 2016 Mw 5.9 Menyuan Earthquake: Implications on Seismogenic Tectonic Motion
by Yongsheng Li, Wenliang Jiang, Jingfa Zhang and Yi Luo
Remote Sens. 2016, 8(6), 519; https://doi.org/10.3390/rs8060519 - 22 Jun 2016
Cited by 45 | Viewed by 7323
Abstract
Determining the relationship between crustal movement and faulting in thrust belts is essential for understanding the growth of geological structures and addressing the proposed models of a potential earthquake hazard. A Mw 5.9 earthquake occurred on 21 January 2016 in Menyuan, NE Qinghai [...] Read more.
Determining the relationship between crustal movement and faulting in thrust belts is essential for understanding the growth of geological structures and addressing the proposed models of a potential earthquake hazard. A Mw 5.9 earthquake occurred on 21 January 2016 in Menyuan, NE Qinghai Tibetan plateau. We combined satellite interferometry from Sentinel-1A Terrain Observation with Progressive Scans (TOPS) images, historical earthquake records, aftershock relocations and geological data to determine fault seismogenic structural geometry and its relationship with the Lenglongling faults. The results indicate that the reverse slip of the 2016 earthquake is distributed on a southwest dipping shovel-shaped fault segment. The main shock rupture was initiated at the deeper part of the fault plane. The focal mechanism of the 2016 earthquake is quite different from that of a previous Ms 6.5 earthquake which occurred in 1986. Both earthquakes occurred at the two ends of a secondary fault. Joint analysis of the 1986 and 2016 earthquakes and aftershocks distribution of the 2016 event reveals an intense connection with the tectonic deformation of the Lenglongling faults. Both earthquakes resulted from the left-lateral strike-slip of the Lenglongling fault zone and showed distinct focal mechanism characteristics. Under the shearing influence, the normal component is formed at the releasing bend of the western end of the secondary fault for the left-order alignment of the fault zone, while the thrust component is formed at the restraining bend of the east end for the right-order alignment of the fault zone. Seismic activity of this region suggests that the left-lateral strike-slip of the Lenglongling fault zone plays a significant role in adjustment of the tectonic deformation in the NE Tibetan plateau. Full article
(This article belongs to the Special Issue Earth Observations for Geohazards)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>(<b>a</b>) Tectonic background of the 21 January 2016 Menyuan Earthquake superimposed on topographic relief. The star is location of the 2016 Menyuan event. The red lines denote the active faults. The blue frames are the coverage of the Sentinel-1A data. The red dots show the historic events since 1927; (<b>b</b>) The partially enlarged view of the black dotted frames in (<b>a</b>). F1: The main fault of Lenglongling; F2: The secondary fault of Lenglongling; the circles express the aftershocks location. Both ends of the secondary fault of Lenglongling are bent to converge to the main fault.</p> Full article ">Figure 2
<p>The light of sight (LOS) deformation maps of the 2016 Menyuan earthquake. Each map is labeled and has a background of shaded topography. The black lines indicates the main and the secondary fault of Lenglongling. (<b>a</b>) Ascending LOS deformation map with pairs 13 January 2016–6 February 2016; (<b>b</b>) Descending LOS deformation map with pairs 18 January2016–11 February 2016. The main and the secondary faults of Lenglongling are labeled consistently with <a href="#remotesensing-08-00519-f001" class="html-fig">Figure 1</a>b.</p> Full article ">Figure 3
<p>Contour map of log function with variations in dips and smoothing coefficients (α<sup>2</sup>). The red star indicates the global minimum point.</p> Full article ">Figure 4
<p>Original, modeled and residual interferograms for InSAR-derived slip models. (<b>a</b>–<b>c</b>) are from SENTINEL-1A Path 128 while (<b>d</b>–<b>f</b>) are from SENTINEL-1A Path 33.</p> Full article ">Figure 5
<p>Slip distribution of the Menyuan earthquake. (<b>a</b>) Coseismic slip model of the 2016 Menyuan earthquake derived using two paths of Sentinel-1A IW deformation maps; (<b>b</b>) The seismic moment distribution along the depth in the slip model.</p> Full article ">Figure 6
<p>Three-dimensional illustration of Slip distribution of the Menyuan earthquake. The blue points represent the relocation of aftershocks, the red star denotes the epicenter.</p> Full article ">Figure 7
<p>Tectonic transformation mode in the Lenglongling fault.</p> Full article ">Figure 8
<p>Three-dimensional block diagram of the geometry proposed for the Lenglongling fault. Red rectangles indicate the compression regions of the secondary fault of Lenglongling, while the blue ones indicate the tensile regions.</p> Full article ">
11555 KiB  
Article
An Automated Approach for Sub-Pixel Registration of Landsat-8 Operational Land Imager (OLI) and Sentinel-2 Multi Spectral Instrument (MSI) Imagery
by Lin Yan, David P. Roy, Hankui Zhang, Jian Li and Haiyan Huang
Remote Sens. 2016, 8(6), 520; https://doi.org/10.3390/rs8060520 - 21 Jun 2016
Cited by 105 | Viewed by 14835
Abstract
Moderate spatial resolution satellite data from the Landsat-8 OLI and Sentinel-2A MSI sensors together offer 10 m to 30 m multi-spectral reflective wavelength global coverage, providing the opportunity for improved combined sensor mapping and monitoring of the Earth’s surface. However, the standard geolocated [...] Read more.
Moderate spatial resolution satellite data from the Landsat-8 OLI and Sentinel-2A MSI sensors together offer 10 m to 30 m multi-spectral reflective wavelength global coverage, providing the opportunity for improved combined sensor mapping and monitoring of the Earth’s surface. However, the standard geolocated Landsat-8 OLI L1T and Sentinel-2A MSI L1C data products are currently found to be misaligned. An approach for automated registration of Landsat-8 OLI L1T and Sentinel-2A MSI L1C data is presented and demonstrated using contemporaneous sensor data. The approach is computationally efficient because it implements feature point detection across four image pyramid levels to identify a sparse set of tie-points. Area-based least squares matching around the feature points with mismatch detection across the image pyramid levels is undertaken to provide reliable tie-points. The approach was assessed by examination of extracted tie-point spatial distributions and tie-point mapping transformations (translation, affine and second order polynomial), dense-matching prediction-error assessment, and by visual registration assessment. Two test sites over Cape Town and Limpopo province in South Africa that contained cloud and shadows were selected. A Landsat-8 L1T image and two Sentinel-2A L1C images sensed 16 and 26 days later were registered (Cape Town) to examine the robustness of the algorithm to surface, atmosphere and cloud changes, in addition to the registration of a Landsat-8 L1T and Sentinel-2A L1C image pair sensed 4 days apart (Limpopo province). The automatically extracted tie-points revealed sensor misregistration greater than one 30 m Landsat-8 pixel dimension for the two Cape Town image pairs, and greater than one 10 m Sentinel-2A pixel dimension for the Limpopo image pair. Transformation fitting assessments showed that the misregistration can be effectively characterized by an affine transformation. Hundreds of automatically located tie-points were extracted and had affine-transformation root-mean-square error fits of approximately 0.3 pixels at 10 m resolution and dense-matching prediction errors of similar magnitude. These results and visual assessment of the affine transformed data indicate that the methodology provides sub-pixel registration performance required for meaningful Landsat-8 OLI and Sentinel-2A MSI data comparison and combined data applications. Full article
(This article belongs to the Special Issue First Experiences with European Sentinel-2 Multi-Spectral Imager (MSI))
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Cape Town, South Africa, test data showing (<b>a</b>) Landsat-8-L1T sensed 22 November 2015 (week 47); (<b>b</b>) Sentinel-2A L1C sensed 8 December 2015 (week 49); and (<b>c</b>) Sentinel-2A L1C sensed 18 December 2015 (week 51). The NIR (Sentinel-2: 842 nm and Landsat-8 864 nm band) TOA reflectance for each image is shown, which was reprojected to 10 m global WELD tile hh19vv12.h3v2 (sinusoidal projection, 15885 × 15885 10 m pixels).</p> Full article ">Figure 2
<p>Limpopo Province, South Africa, test data showing (<b>a</b>) Landsat-8-L1T sensed 5 December 2015 (week 49); (<b>b</b>) Sentinel-2A L1C sensed 9 December 2015 (week 49). The NIR TOA reflectance for each image is shown, which was reprojected to 10 m global WELD tile hh20vv11.h4v3 (sinusoidal projection, 15885 × 15885 10 m pixels).</p> Full article ">Figure 3
<p>Automated workflow to register Landsat-8 OLI to Sentinel-2A MSI WELD tiles.</p> Full article ">Figure 4
<p>Illustration of depth-first LSM mismatch detection on the four-level image pyramid (shown for the Sentinel-2A image only).</p> Full article ">Figure 5
<p>Illustration of the tie-points and misregistration vectors obtained from the Cape Town data (<a href="#remotesensing-08-00520-f001" class="html-fig">Figure 1</a>). The 116 green vectors point from the Landsat-8 week 47 image tie-point locations to the corresponding Sentinel-2A week 49 tie point locations. The 797 red vectors point from the Landsat-8 week 47 image tie-point locations to the corresponding Sentinel-2A week 51 tie-point locations. The vector lengths are enlarged by 80 times for visual clarity. To provide geographic context, the background image shows the Landsat-8 week 47 30 m true color image.</p> Full article ">Figure 6
<p>False color images illustrating Landsat-8 week 47 and Sentinel-2A week 51 images (<b>a</b>) before registration and (<b>b</b>) after registration. The Sentinel NIR data are shown as red and the Landsat NIR data are shown as blue and green. A 350 × 350 30 m pixel subset over Saldanha Bay, Cape Town (northern side of the WELD tile, <a href="#remotesensing-08-00520-f001" class="html-fig">Figure 1</a>) is shown. The registration was undertaken using the affine transformation coefficients (<a href="#remotesensing-08-00520-t003" class="html-table">Table 3</a>).</p> Full article ">Figure 7
<p>Dense-matching prediction-error maps for the translation (<b>a</b>); affine (<b>b</b>) and second order polynomial (<b>c</b>) transformations between the Cape Town Landsat-8 week 47 and Sentinel-2A week 51 image pair. Dense-matching grid points were sampled every six 10 m pixel across the 15885 × 15885 10 m WELD tile, generating 2647 × 2647 prediction-error maps (Equation (7)). Locations where there are no matches are shown as black.</p> Full article ">Figure 8
<p>Illustration of the tie-points and misregistration vectors obtained from the Limpopo data (<a href="#remotesensing-08-00520-f002" class="html-fig">Figure 2</a>). The vectors point from the Landsat-8 week 49 image tie-point locations to the Sentinel-2A week 49 (180 red vectors) tie-point locations. The vector lengths are enlarged by 200 times for visual clarity. To provide geographic context the background image shows the Landsat-8 week 49 30 m true color image.</p> Full article ">Figure 9
<p>Dense-matching prediction-error maps for the translation (<b>a</b>); affine (<b>b</b>); and second order polynomial (<b>c</b>) transformations, between the Limpopo Landsat-8 week 49 and Sentinel-2A week 49 image pair.</p> Full article ">Figure 10
<p>Dense-matching maps x and y axis shifts (units 10 m pixels) between the Limpopo Landsat-8 week 49 and Sentinel-2A week 49 image pair. The affine transformation was used to guide the dense matching (<a href="#sec4dot4-remotesensing-08-00520" class="html-sec">Section 4.4</a>). The (<b>a</b>) x-shift <math display="inline"> <semantics> <mrow> <msub> <mi>x</mi> <mrow> <mi>S</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mo>−</mo> <msub> <mi>x</mi> <mrow> <mi>L</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> </mrow> </semantics></math> and (<b>b</b>) y-shift <math display="inline"> <semantics> <mrow> <msub> <mi>y</mi> <mrow> <mi>L</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mo>−</mo> <msub> <mi>y</mi> <mrow> <mi>L</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> </mrow> </semantics> </math> are shown, where (<math display="inline"> <semantics> <mrow> <msub> <mi>x</mi> <mrow> <mi>S</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mtext> </mtext> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>S</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> </mrow> </semantics> </math>) is the Sentinel grid-point location and (<math display="inline"> <semantics> <mrow> <msub> <mi>x</mi> <mrow> <mi>L</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mtext> </mtext> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>L</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> </mrow> </semantics> </math>) is the corresponding least squares matched Landsat location for grid-point <span class="html-italic">i</span>. Locations where there are no matches are shown as black. Note that (<math display="inline"> <semantics> <mrow> <msub> <mi>x</mi> <mrow> <mi>L</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mtext> </mtext> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>L</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> </mrow> </semantics> </math>) is theoretically independent on the transformation type and so the translation and polynomial-based shift maps, which were very similar to the affine-based shift maps, are not shown; (<b>c</b>) shows the Sentinel-2A L1C tile and detector boundaries (red) and (<b>d</b>) shows the Landsat-8 L1T image boundaries (blue).</p> Full article ">
5662 KiB  
Article
Merging Alternate Remotely-Sensed Soil Moisture Retrievals Using a Non-Static Model Combination Approach
by Seokhyeon Kim, Robert M. Parinussa, Yi Y. Liu, Fiona M. Johnson and Ashish Sharma
Remote Sens. 2016, 8(6), 518; https://doi.org/10.3390/rs8060518 - 21 Jun 2016
Cited by 15 | Viewed by 5980
Abstract
Soil moisture is an important variable in the coupled hydrologic and climate system. In recent years, microwave-based soil moisture products have been shown to be a viable alternative to in situ measurements. A popular way to measure the performance of soil moisture products [...] Read more.
Soil moisture is an important variable in the coupled hydrologic and climate system. In recent years, microwave-based soil moisture products have been shown to be a viable alternative to in situ measurements. A popular way to measure the performance of soil moisture products is to calculate the temporal correlation coefficient (R) against in situ measurements or other appropriate reference datasets. In this study, an existing linear combination method improving R was modified to allow for a non-static or nonstationary model combination as the basis for improving remotely-sensed surface soil moisture. Previous research had noted that two soil moisture products retrieved using the Japan Aerospace Exploration Agency (JAXA) and Land Parameter Retrieval Model (LPRM) algorithms from the same Advanced Microwave Scanning Radiometer 2 (AMSR2) sensor are spatially complementary in terms of R against a suitable reference over a fixed period. Accordingly, a linear combination was proposed to maximize R using a set of spatially-varying, but temporally-fixed weights. Even though this approach showed promising results, there was room for further improvements, in particular using non-static or dynamic weights that take account of the time-varying nature of the combination algorithm being approximated. The dynamic weighting was achieved by using a moving window. A number of different window sizes was investigated. The optimal weighting factors were determined for the data lying within the moving window and then used to dynamically combine the two parent products. We show improved performance for the dynamically-combined product over the static linear combination. Generally, shorter time windows outperform the static approach, and a 60-day time window is suggested to be the optimum. Results were validated against in situ measurements collected from 124 stations over different continents. The mean R of the dynamically-combined products was found to be 0.57 and 0.62 for the cases using the European Centre for Medium-Range Weather Forecasts Reanalysis-Interim (ERA-Interim) and Modern-Era Retrospective Analysis for Research and Applications Land (MERRA-Land) reanalysis products as the reference, respectively, outperforming the statically-combined products (0.55 and 0.54). Full article
(This article belongs to the Special Issue Multi-Sensor and Multi-Data Integration in Remote Sensing)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Locations of 124 ground stations from 10 networks used for the comparison with combined products.</p> Full article ">Figure 2
<p>Schematic diagram for dynamic linear combination. T denotes the period defined by the window (<span class="html-italic">i.e</span>., T = (t − N/2):(t + N/2)). Therefore, a bold symbol that has T as its subscript means a vector in the period T, and a non-bold symbol with t as the subscript represents a value at the point in time <span class="html-italic">t</span>.</p> Full article ">Figure 3
<p>Results from experiments that uses ERA-Interim as the reference for various window sizes, N60, N90 and N120. Each panel shows the R between the reference and (<b>a</b>) JAXA; (<b>b</b>) LPRM; (<b>c</b>) static; (<b>d</b>) N60; (<b>e</b>) N90 and (<b>f</b>) N120; the more bluish colours in the maps indicate higher R against the reference; the overall performance for the various scenarios is summarized in the boxplot (<b>g</b>).</p> Full article ">Figure 4
<p>Comparison between combined soil moisture products. For ERA-Interim as the reference, (<b>a</b>) The differences in R between the static and N60 products against the reference (<span class="html-italic">i.e</span>., R of N60 minus R of static) and (<b>b</b>) the mean weights that were used for the dynamic combination using the reference over the two-year study period; (<b>c</b>) and (<b>d</b>) show corresponding results with (<b>a</b>) and (<b>b</b>) when using MERRA-Land as the reference.</p> Full article ">Figure 5
<p>Results from the simulation experiment. (<b>a</b>) The x-axis indicates Euclidean distances (ξ) calculated by Equation (7), representing the qualities of the parent products, and the y-axis, <math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mrow> <mi>d</mi> <mi>y</mi> <mi>n</mi> </mrow> </msub> </mrow> </semantics> </math> or <math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mrow> <mi>s</mi> <mi>t</mi> <mi>a</mi> </mrow> </msub> </mrow> </semantics> </math>. The dashed two lines present the linear regression of all results from the dynamic and static combinations, respectively; (<b>b</b>) The x-axis indicates N sizes, the y-axis differences between <math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mrow> <mi>d</mi> <mi>y</mi> <mi>n</mi> </mrow> </msub> </mrow> </semantics> </math> and <math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mrow> <mi>s</mi> <mi>t</mi> <mi>a</mi> </mrow> </msub> </mrow> </semantics> </math> (<span class="html-italic">i.e</span>., <math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mrow> <mi>d</mi> <mi>y</mi> <mi>n</mi> </mrow> </msub> </mrow> </semantics> </math> − <math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mrow> <mi>s</mi> <mi>t</mi> <mi>a</mi> </mrow> </msub> </mrow> </semantics> </math>).</p> Full article ">Figure 6
<p>(<b>a</b>) Box plots showing combination performances against <span class="html-italic">in situ</span> measurements with the N60 and the two references. The labels on the x-axis indicate parent or statically-/dynamically-combined products with the references, and the y-axis R between the product and the <span class="html-italic">in situ</span> measurements. The value in each box is the mean of R. Comparison against <span class="html-italic">in situ</span> measurements from the ISMN for dynamically combined products using the N60 and (<b>b</b>) ERA-Interim and (<b>c</b>) MERRA-Land as the reference, respectively. The x-axis presents R between a dynamic product and the <span class="html-italic">in situ</span> measurements from a station, the y-axis R between a static product and the <span class="html-italic">in situ</span> measurements.</p> Full article ">Figure 7
<p>Dynamic and static combination results using MERRA-Land as the reference at (<b>a</b>) Sandy Ridge station in Soil Climate Analysis Network and (<b>b</b>) Sandstone-6-W station in U.S. Climate Reference Network. Each panel shows static/dynamic weights (<b>top</b>), as well as time series of statically- and dynamically-combined soil moisture products (<b>bottom</b>).</p> Full article ">Figure 8
<p>Combination performances with the quality of parent products and reference against <span class="html-italic">in situ</span> measurements. (<b>a</b>) ERA-Interim; (<b>b</b>) MERRA-Land. The x-axis for each panel presents the Euclidean distances (ξ) calculated by Equation (8), and the y-axis <math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mrow> <mi>i</mi> <mi>n</mi> <mi>s</mi> <mi>i</mi> <mi>t</mi> <mi>u</mi> <mo>−</mo> <mi>s</mi> <mi>t</mi> <mi>a</mi> </mrow> </msub> </mrow> </semantics> </math> or <math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mrow> <mi>i</mi> <mi>n</mi> <mi>s</mi> <mi>i</mi> <mi>t</mi> <mi>u</mi> <mo>−</mo> <mi>d</mi> <mi>y</mi> <mi>n</mi> </mrow> </msub> </mrow> </semantics> </math>. Linear regression lines represent the tendencies of both cases.</p> Full article ">Figure 9
<p>Combination performances with reference quality against <span class="html-italic">in situ</span> measurements. The x-axis presents R between <span class="html-italic">in situ</span> measurements and the references (<math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mrow> <mi>i</mi> <mi>n</mi> <mi>s</mi> <mi>i</mi> <mi>t</mi> <mi>u</mi> <mo>−</mo> <mi>r</mi> <mi>e</mi> <mi>f</mi> </mrow> </msub> </mrow> </semantics> </math>), the y-axis R between <span class="html-italic">in situ</span> measurements and statically-/dynamically-combined products (<math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mrow> <mi>i</mi> <mi>n</mi> <mi>s</mi> <mi>i</mi> <mi>t</mi> <mi>u</mi> <mo>−</mo> <mi>s</mi> <mi>t</mi> <mi>a</mi> </mrow> </msub> </mrow> </semantics> </math> and <math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mrow> <mi>i</mi> <mi>n</mi> <mi>s</mi> <mi>i</mi> <mi>t</mi> <mi>u</mi> <mo>−</mo> <mi>d</mi> <mi>y</mi> <mi>n</mi> </mrow> </msub> </mrow> </semantics> </math>). Linear regression lines are added for representing the average tendencies of both cases.</p> Full article ">
5714 KiB  
Article
Developing a Comprehensive Spectral-Biogeochemical Database of Midwestern Rivers for Water Quality Retrieval Using Remote Sensing Data: A Case Study of the Wabash River and Its Tributary, Indiana
by Jing Tan, Keith A. Cherkauer and Indrajeet Chaubey
Remote Sens. 2016, 8(6), 517; https://doi.org/10.3390/rs8060517 - 21 Jun 2016
Cited by 13 | Viewed by 6884
Abstract
A comprehensive spectral-biogeochemical database was developed for the Wabash River and the Tippecanoe River in Indiana, United States. This database includes spectral measurements of river water, coincident in situ measurements of water quality parameters (chlorophyll (chl), non-algal particles (NAP), and colored dissolved organic [...] Read more.
A comprehensive spectral-biogeochemical database was developed for the Wabash River and the Tippecanoe River in Indiana, United States. This database includes spectral measurements of river water, coincident in situ measurements of water quality parameters (chlorophyll (chl), non-algal particles (NAP), and colored dissolved organic matter (CDOM)), nutrients (total nitrogen (TN), total phosphorus (TP), and dissolved organic carbon (DOC)), water-column inherent optical properties (IOPs), water depths, substrate types, and bottom reflectance spectra collected in summer 2014. With this dataset, the temporal variability of water quality observations was first analyzed and studied. Second, radiative transfer models were inverted to retrieve water quality parameters using a look-up table (LUT) based spectrum matching methodology. Results found that the temporal variability of water quality parameters and nutrients in the Wabash River was closely associated with hydrologic conditions. Meanwhile, there were no significant correlations found between these parameters and streamflow for the Tippecanoe River, due to the two upstream reservoirs, which increase the settling of sediment and uptake of nutrients. The poor relationship between CDOM and DOC indicates that most DOC in the rivers was from human sources such as wastewater. It was also found that the source of water (surface runoff or combined sewer overflow (CSO)), water temperature, and nutrients were important factors controlling instream concentrations of phytoplankton. The LUT retrieved NAP concentrations were in good agreement with field measurements with slope close to 1.0 and the average estimation error was 4.1% of independently obtained lab measurements. The error for chl estimation was larger (37.7%), which is attributed to the fact that the specific absorption spectrum of chl was not well represented in this study. The LUT retrievals for CDOM experienced large variability, probably due to the small data range collected in this study and the insensitivity of Rrs to CDOM change. It is concluded that the success of the LUT method requires accurate spectral measurements and enough a priori information of the environment to construct a representative database for water quality retrieval. Therefore, future work will focus on continuing data collection in other seasons of the year and better characterization of the study area. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Main study area includes two reaches of the Wabash River, including the confluence with the Tippecanoe River. Field spectrometer measurements and water samples (marked as red stars) were collected through the summer of 2014. Triangles indicate United Stated Geological Survey (USGS) real-time streamflow monitoring stations.</p> Full article ">Figure 2
<p>Scatterplots of measured (<b>a</b>) concentrations of total suspended sediments ((TSS)] and (<b>b</b>) concentrations of colored dissolved organic matter (<span class="html-italic">a</span><sub>cdom</sub>(440)) <span class="html-italic">versus</span> chlorophyll concentrations ((chl)) for samples in the Wabash River (circles) and the Tippecanoe River (triangles) in summer 2014.</p> Full article ">Figure 3
<p>Time series of measured concentrations of water quality parameters and nutrients (circles) <span class="html-italic">versus</span> streamflow (solid line) of: (<b>a</b>) the Wabash River; and (<b>b</b>) the Tippecanoe River.</p> Full article ">Figure 4
<p>Scatterplot showing measured DOC concentrations ((DOC)) <span class="html-italic">versus</span> concentrations of colored dissolved organic matter (<span class="html-italic">a</span><sub>cdom</sub>(440)) for samples collected for the Wabash River (circles) and the Tippecanoe River (triangles) in summer 2014.</p> Full article ">Figure 5
<p>Boxplots of measured chlorophyll concentrations ((chl)) for samples collected in the Wabash River during the summer of 2014. Boxes filled with yellow color indicate that there is a statistically significant different (<span class="html-italic">p</span> &lt; 0.05) between observations on the highlighted day and the previous day.</p> Full article ">Figure 6
<p>Scatterplots of measured inherent optical properties (filled squares) <span class="html-italic">versus</span> water quality parameters (circles) of the Wabash River (left panel) and the Tippecanoe River (right panel): (<b>a</b>,<b>b</b>)—absorption of phytoplankton at 676 nm (<span class="html-italic">a</span><sub>ph</sub>(676)) <span class="html-italic">vs.</span> chlorophyll concentrations ((chl)); (<b>c</b>,<b>d</b>)—absorption of non-algal particles at 440 nm (<span class="html-italic">a</span><sub>nap</sub>(440)) <span class="html-italic">vs.</span> concentrations of total suspended sediments ((TSS)).</p> Full article ">Figure 7
<p>Specific inherent optical properties for the Wabash River and the Tippecanoe River: (<b>a</b>) absorptions; and (<b>b</b>) backscattering.</p> Full article ">Figure 8
<p>Variability of the retrieved backscattering coefficients of particles at 550 nm (<span class="html-italic">b</span><sub>b,p</sub>(550), filled squares) with measured (<b>a</b>) concentrations of total suspended sediments ((TSS)) (circles) and (<b>b</b>) concentrations of chlorophyll ((chl)) (circles) in the Wabash River and the Tippecanoe River.</p> Full article ">Figure 9
<p>Bottom types identified for the Wabash River and the Tippecanoe River: (<b>a</b>) fines; (<b>b</b>) sand; (<b>c</b>) gravel; and (<b>d</b>) cobble.</p> Full article ">Figure 10
<p>Albedo measured for different bottom types of the Wabash River and the Tippecanoe River.</p> Full article ">Figure 11
<p>Examples showing phytoplankton dominated (solid lines) and sediment dominated (dotted line) spectra.</p> Full article ">Figure 12
<p>Comparison between test and look-up table estimated concentrations of: (<b>a</b>) non-algal particles ((NAP)); (<b>b</b>) chlorophyll ((chl)); and (<b>c</b>) colored organic matter (<span class="html-italic">a</span><sub>cdom</sub>(440)), for the Wabash River and the Tippecanoe River. The dotted line represents 1:1 line.</p> Full article ">Figure 13
<p>Comparison between measured and look-up table estimated (unconstrained inversion) concentrations of: (<b>a</b>) non-algal particles ((NAP)); (<b>b</b>) chlorophyll ((chl)); and (<b>c</b>) colored organic matter (<span class="html-italic">a</span><sub>cdom</sub>(440)) for the Wabash River and the Tippecanoe River. The dotted line represents 1:1 line and the dashed lines represent 95% confidence interval.</p> Full article ">Figure 13 Cont.
<p>Comparison between measured and look-up table estimated (unconstrained inversion) concentrations of: (<b>a</b>) non-algal particles ((NAP)); (<b>b</b>) chlorophyll ((chl)); and (<b>c</b>) colored organic matter (<span class="html-italic">a</span><sub>cdom</sub>(440)) for the Wabash River and the Tippecanoe River. The dotted line represents 1:1 line and the dashed lines represent 95% confidence interval.</p> Full article ">Figure 14
<p>Comparison between measured and look-up table estimated (unconstrained inversion) values of chlorophyll concentrations ((chl)) for selected samples showing fewer features of accessory pigments. The dotted line represents 1:1 line and the dashed lines represent 95% confidence interval.</p> Full article ">
40326 KiB  
Article
Source Parameters of the 2003–2004 Bange Earthquake Sequence, Central Tibet, China, Estimated from InSAR Data
by Lingyun Ji, Jing Xu, Qiang Zhao and Chengsheng Yang
Remote Sens. 2016, 8(6), 516; https://doi.org/10.3390/rs8060516 - 18 Jun 2016
Cited by 7 | Viewed by 6852
Abstract
A sequence of Ms ≥ 5.0 earthquakes occurred in 2003 and 2004 in Bange County, Tibet, China, all with similar depths and focal mechanisms. However, the source parameters, kinematics and relationships between these earthquakes are poorly known because of their moderately-sized magnitude and [...] Read more.
A sequence of Ms ≥ 5.0 earthquakes occurred in 2003 and 2004 in Bange County, Tibet, China, all with similar depths and focal mechanisms. However, the source parameters, kinematics and relationships between these earthquakes are poorly known because of their moderately-sized magnitude and the sparse distribution of seismic stations in the region. We utilize interferometric synthetic aperture radar (InSAR) data from the European Space Agency’s Envisat satellite to determine the location, fault geometry and slip distribution of three large events of the sequence that occurred on 7 July 2003 (Ms 6.0), 27 March 2004 (Ms 6.2), and 3 July 2004 (Ms 5.1). The modeling results indicate that the 7 July 2003 event was a normal-faulting event with a right-lateral slip component, the 27 March 2004 earthquake was associated with a normal fault striking northeast–southwest and dipping northwest with a moderately oblique right-lateral slip, and the 3 July 2004 event was caused by a normal fault. A calculation of the static stress changes on the fault planes demonstrates that the third earthquake may have been triggered by the previous ones. Full article
(This article belongs to the Special Issue Earth Observations for Geohazards)
►▼ Show Figures

Figure 1

Figure 1
<p>Topographic map of Bange County in central Tibet, China, with the location shown in the inset. Green lines in inset represent block boundaries [<a href="#B3-remotesensing-08-00516" class="html-bibr">3</a>]: BB, Bayan Har Block; QB, Qiangtang Block. Shaded relief topography is SRTM DEM at 90 m resolution. Black thin lines are fault traces [<a href="#B4-remotesensing-08-00516" class="html-bibr">4</a>]. Earthquakes listed in <a href="#remotesensing-08-00516-t001" class="html-table">Table 1</a> are shown as red circles. Blue circles are aftershocks with Ms ≥ 3.0 through 2015. Earthquake catalogue is from China Earthquake Networks Center (CENC) [<a href="#B1-remotesensing-08-00516" class="html-bibr">1</a>]. Black box with solid line marks areas covered by interferograms of the 7 July 2003 event. Dashed box marks areas covered by interferograms of the 27 March 2004 event. Green box marks areas covered by interferograms of the 3 July 2004 event. Focal mechanisms from NEIC and GCMT for 7 July 2003 Ms 6.0, 27 March 2004 Ms 6.2, and 3 July 2004 Ms 5.1 events are shown.</p> Full article ">Figure 2
<p>Time intervals covered by each of the interferograms shown in <a href="#remotesensing-08-00516-f003" class="html-fig">Figure 3</a> and <a href="#remotesensing-08-00516-f004" class="html-fig">Figure 4</a>. Values in parentheses are the perpendicular baseline of each interferogram. Grey bars show time intervals. Black solid vertical lines mark the times of the 7 July 2003 Ms 6.0, 27 March 2004 Ms 6.2, and 3 July 2004 earthquakes (see <a href="#remotesensing-08-00516-t001" class="html-table">Table 1</a> for details).</p> Full article ">Figure 3
<p>Coseismic interferograms of the 7 July 2003 earthquake. Location of the interferograms is shown in <a href="#remotesensing-08-00516-f001" class="html-fig">Figure 1</a> as a box with solid line. Start and end dates are provided above each image using the format yyyymmdd. (<b>a</b>) 20030409–20040114. Satellite flight direction and radar look direction are labeled as a solid arrow and open arrow, respectively; (<b>b</b>) 20030618–20030723. Each fringe, <span class="html-italic">i.e.</span>, full color cycle from red through yellow to blue, represents 28 mm of range increase between the ground and the satellite.</p> Full article ">Figure 4
<p>Interferograms related to the 2004 earthquakes. Location of the interferograms is shown in <a href="#remotesensing-08-00516-f001" class="html-fig">Figure 1</a> as a dashed box. Start and end dates are provided above each image using the format yyyymmdd. White solid arrows point to the oval pattern caused by the 27 March 2004 event, whereas the white dashed circles delineate the circular pattern caused by the 3 July 2004 event. Satellite flight direction and radar look direction are labeled as short solid arrow and open arrow, respectively. Each fringe, <span class="html-italic">i.e.</span>, full color cycle from red through yellow to blue, represents 28 mm of range increase between the ground and satellite. (<b>a</b>) 20030409–20040114; (<b>b</b>) 20030409–20041124; (<b>c</b>) 20030618–20050622; (<b>d</b>) 20040114–20040622; (<b>e</b>) 20040114–20041124; (<b>f</b>) 20040114–20050622; (<b>g</b>) 20040602–20050622; (<b>h</b>) 20041124–20070207.</p> Full article ">Figure 5
<p>Frequency histograms of modelled parameters determined from 1000 independent runs of the inversion algorithm. Histograms represent the 1000 best-fit solution parameters (<b>black</b> bins) obtained from inversions of InSAR coseismic deformation maps. The optimal solution for the parameters is estimated from the mean value (<b>blue</b> vertical line) of the best-fit Gaussian (<b>red</b> curve). (<b>a</b>) 7 July 2003 earthquake; (<b>b</b>) 3 July 2004 earthquake; (<b>c</b>) 27 March 2004 earthquake.</p> Full article ">Figure 6
<p>Coseismic deformation (range displacement–negative away from the satellite) and model for uniform slip inversion of the 7 July 2003 earthquake. (<b>a</b>) Observed interferogram spanning 20030618–20030723. Black and blue beach balls show focal mechanisms of NEIC and GCMT catalogues, respectively; (<b>b</b>) Synthetic interferogram for a uniform slip elastic dislocation model; (<b>c</b>) Residual interferogram, which is the difference between observed (<b>a</b>) and modeled (<b>b</b>) interferograms; (<b>d</b>,<b>e</b>,<b>f</b>) are profiles of line-of-sight (LOS) displacements (<b>blue</b> dots), model LOS displacements (<b>red</b> dots) and topography (<b>grey</b>), respectively. Crosses in (<b>a</b>) indicate profile locations. Black line in (<b>a</b>) represents the modeled fault trace.</p> Full article ">Figure 7
<p>Coseismic deformation (range displacement—negative away from the satellite) and model for uniform slip inversion of the 3 July 2004 earthquake. Location of the interferograms is shown in <a href="#remotesensing-08-00516-f001" class="html-fig">Figure 1</a> as a green box. (<b>a</b>) Observed interferogram spanning 20040602–20050622. Focal mechanism from GCMT catalogue is shown; (<b>b</b>) Synthetic interferogram for a uniform slip elastic dislocation model; (<b>c</b>) Residual interferogram, which is the difference between observed (<b>a</b>) and modeled (<b>b</b>) interferograms; (<b>d</b>) Profile of line-of-sight (LOS) displacements (<b>blue</b> dots), model LOS displacements (<b>red</b> dots) and topography (<b>grey</b>). Crosses in (<b>a</b>) indicate profile locations. Black lines in (<b>a</b>) represent modeled fault trace.</p> Full article ">Figure 8
<p>Coseismic deformation (range displacement—negative away from the satellite) and model for uniform slip inversion of the 27 March 2004 earthquake. (<b>a</b>) Observed interferogram spanning 20040114–20041124. Black and blue beach balls show focal mechanisms from NEIC and GCMT catalogues, respectively; (<b>b</b>) Observed interferogram spanning 20040114–20041124 obtained by subtracting <a href="#remotesensing-08-00516-f007" class="html-fig">Figure 7</a>b; (<b>c</b>) Synthetic interferogram for uniform slip elastic dislocation model; (<b>d</b>) Residual interferogram, which is the difference between observed (<b>b</b>) and modeled (<b>c</b>) interferograms; (<b>e</b>) Profile of line-of-sight (LOS) displacements (<b>blue</b> dots), model LOS displacements (<b>red</b> dots), and topography (<b>grey</b>). Crosses in (<b>a</b>) indicate profile locations. Black line in (<b>a</b>) represents modeled fault trace.</p> Full article ">Figure 9
<p>Trade-off curves between misfit and model roughness. The roughness is the normalized value. Pluses indicate locations of optimal smoothing parameters where balances between model misfit and smoothness is achieved. (<b>a</b>) 7 July 2003 earthquake; (<b>b</b>) 3 July 2004 earthquake; (<b>c</b>) 27 March 2004 earthquake.</p> Full article ">Figure 10
<p>Coseismic deformation (range displacement—negative away from the satellite) and model for distributed slip inversion for the 7 July 2003 earthquake. (<b>a</b>) Observed interferogram spanning 20030618–20030723. Black and blue beach balls show focal mechanisms from NEIC and GCMT catalogues, respectively; (<b>b</b>) Synthetic interferogram and (<b>c</b>) residual interferogram based upon the fault plane (<b>black</b> line in (<b>a</b>)) slip distribution shown in <a href="#remotesensing-08-00516-f011" class="html-fig">Figure 11</a>; (<b>d</b>,<b>e</b>,<b>f</b>) are profiles of line-of-sight (LOS) displacements (<b>blue</b> dots), model LOS displacements (<b>red</b> dots), and topography (<b>grey</b>), respectively. Crosses in (<b>a</b>) indicate profile locations.</p> Full article ">Figure 11
<p>Slip distributions for modeled seismic source of the 7 July 2003 event. (<b>a</b>) Perpendicular view of the fault, with slip vectors plotted in addition to the slip magnitudes shown in color; (<b>b</b>) 3-D view from WSW; (<b>c</b>) 1σ uncertainty for slip distribution as shown in (<b>a</b>) and (<b>b</b>), estimated from performing 100 inversions.</p> Full article ">Figure 12
<p>Coseismic deformation (range displacement—negative away from the satellite) and model for distributed slip inversion of the 3 July 2004 earthquake. (<b>a</b>) Observed interferogram spanning 20040602–20050622. Focal mechanism from GCMT catalogue is shown; (<b>b</b>) Synthetic interferogram and (<b>c</b>) residual interferogram based upon the fault plane’ (black line in (<b>a</b>)) slip distribution shown in <a href="#remotesensing-08-00516-f013" class="html-fig">Figure 13</a>; (<b>d</b>) Profile of line-of-sight (LOS) displacements (<b>blue</b> dots), model LOS displacements (<b>red</b> dots), and topography (<b>grey</b>). Crosses in (<b>a</b>) indicate profile locations.</p> Full article ">Figure 13
<p>Slip distributions for the modeled seismic sources of the 3 July 2004 event. (<b>a</b>) Perpendicular view of the fault, with slip vectors plotted in addition to the slip magnitudes shown in color; (<b>b</b>) 3-D view from ENE; (<b>c</b>) 1σ uncertainty for the slip distribution as shown in (<b>a</b>,<b>b</b>), estimated from performing 100 inversions.</p> Full article ">Figure 14
<p>Coseismic deformation (range displacement—negative away from the satellite) and model for distributed slip inversion of the 27 March 2004 earthquake. (<b>a</b>) Observed interferogram spanning 20040114–20041124 obtained by subtracting <a href="#remotesensing-08-00516-f004" class="html-fig">Figure 4</a>g. Black and blue beach balls show focal mechanisms from NEIC and GCMT catalogues, respectively; (<b>b</b>) Synthetic interferogram and (<b>c</b>) residual interferogram based upon the fault plane’s (<b>black</b> line in (<b>a</b>)) slip distribution shown in <a href="#remotesensing-08-00516-f015" class="html-fig">Figure 15</a>; (<b>d</b>) Profile of the line-of-sight (LOS) displacements (<b>blue</b> dots), model LOS displacements (<b>red</b> dots), and topography (<b>grey</b>). Crosses in (<b>a</b>) indicate profile locations.</p> Full article ">Figure 15
<p>Slip distribution for modeled seismic source of the 27 March 2004 event. (<b>a</b>) Perpendicular view of the fault, with slip vectors plotted in addition to the slip magnitudes shown in color; (<b>b</b>) 3-D view from WSE; (<b>c</b>) 1σ uncertainty for the slip distribution as shown in (<b>a</b>,<b>b</b>), estimated from performing 100 inversions.</p> Full article ">Figure 16
<p>(<b>a</b>) Coseismic Coulomb stress change on the fault plane of the 27 March 2004 event triggered by the 7 July 2003 event; (<b>b</b>) Stress change induced on the 3 July 2004 earthquake triggered by 7 July 2003 and 27 March 2004 events.</p> Full article ">Figure 17
<p>Normal faulting earthquakes with Mw ≥ 5.5 in Tibetan Plateau (1976–2015). Focal mechanisms are based on GCMT catalogue.</p> Full article ">
5513 KiB  
Article
Large-Area, High-Resolution Tree Cover Mapping with Multi-Temporal SPOT5 Imagery, New South Wales, Australia
by Adrian Fisher, Michael Day, Tony Gill, Adam Roff, Tim Danaher and Neil Flood
Remote Sens. 2016, 8(6), 515; https://doi.org/10.3390/rs8060515 - 18 Jun 2016
Cited by 36 | Viewed by 9701
Abstract
Tree cover maps are used for many purposes, such as vegetation mapping, habitat connectivity and fragmentation studies. Small remnant patches of native vegetation are recognised as ecologically important, yet they are underestimated in remote sensing products derived from Landsat. High spatial resolution sensors [...] Read more.
Tree cover maps are used for many purposes, such as vegetation mapping, habitat connectivity and fragmentation studies. Small remnant patches of native vegetation are recognised as ecologically important, yet they are underestimated in remote sensing products derived from Landsat. High spatial resolution sensors are capable of mapping small patches of trees, but their use in large-area mapping has been limited. In this study, multi-temporal Satellite pour l’Observation de la Terre 5 (SPOT5) High Resolution Geometrical data was pan-sharpened to 5 m resolution and used to map tree cover for the Australian state of New South Wales (NSW), an area of over 800,000 km2. Complete coverages of SPOT5 panchromatic and multispectral data over NSW were acquired during four consecutive summers (2008–2011) for a total of 1256 images. After pre-processing, the imagery was used to model foliage projective cover (FPC), a measure of tree canopy density commonly used in Australia. The multi-temporal imagery, FPC models and 26,579 training pixels were used in a binomial logistic regression model to estimate the probability of each pixel containing trees. The probability images were classified into a binary map of tree cover using local thresholds, and then visually edited to reduce errors. The final tree map was then attributed with the mean FPC value from the multi-temporal imagery. Validation of the binary map based on visually assessed high resolution reference imagery revealed an overall accuracy of 88% (±0.51% standard error), while comparison against airborne lidar derived data also resulted in an overall accuracy of 88%. A preliminary assessment of the FPC map by comparing against 76 field measurements showed a very good agreement (r2 = 0.90) with a root mean square error of 8.57%, although this may not be representative due to the opportunistic sampling design. The map represents a regionally consistent and locally relevant record of tree cover for NSW, and is already widely used for natural resource management in the state. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Examples of native trees present across New South Wales (NSW), Australia. (<b>A</b>) Remnant trees in the cleared agricultural land of the central west; (<b>B</b>) Open woodland in the semi-arid north west; (<b>C</b>) Sclerophyll forest on the east coast; (<b>D</b>) Rainforest in the north east; (<b>E</b>) The locations of each example.</p> Full article ">Figure 2
<p>The method combined automated and manual processing to output the 5 m binary tree cover map and foliage projective cover (FPC) map of New South Wales, Australia. Manual processing steps were assisted by high resolution imagery (0.5–2.5 m).</p> Full article ">Figure 3
<p>The distribution of training and validation datasets used in the development of the New South Wales (NSW) maps. (<b>A</b>) The location of images across NSW and Queensland (QLD) used to train the SPOT5 FPC model using Landsat FPC data; (<b>B</b>) Sites used to train the probability model interpreted from high resolution (0.5–2.5 m) imagery; (<b>C</b>) Sites of field measured FPC, used to validate the SPOT5 FPC map; (<b>D</b>) The sample of airborne lidar taken from the available lidar data used to validate the binary tree cover map; (<b>E</b>) Pixels interpreted from high resolution (2.5 m) imagery used to validate the binary tree cover map, which were stratified by catchment.</p> Full article ">Figure 4
<p>Density plot of foliage projective cover (FPC) derived from multi-temporal SPOT5 imagery compared with that from multi-temporal Landsat imagery, after both maps were resampled to 100 m pixels (darker points have a greater density of observations). The dashed line and equation represent the line of best fit derived from orthogonal distance regression (appropriate when both x and y have error), which shows an excellent agreement with low root mean square error (RMSE) and high correlation (<span class="html-italic">r</span><sup>2</sup>).</p> Full article ">Figure 5
<p>A comparison of the modelled foliage projective cover (FPC) derived from the time-series of SPOT5 imagery to field measured FPC at 76 sites across New South Wales, Australia. The dashed lines and equations represent the line of best fit derived from ordinary least squares regression, which confirmed that the SPOT5 FPC had a stronger relationship (lower RMSE and higher <span class="html-italic">r</span><sup>2</sup>) to field data when the midstorey component was included.</p> Full article ">Figure 6
<p>The final 5 m foliage projective cover (FPC) map for New South Wales, Australia. The locations of the examples shown in <a href="#remotesensing-08-00515-f007" class="html-fig">Figure 7</a> are labelled (<b>A</b>–<b>D</b>).</p> Full article ">Figure 7
<p>Examples of the final 5 m foliage projective cover (FPC) map compared to pan-sharpened false colour SPOT5 imagery. (<b>A</b>) Remnant trees in a cleared agricultural landscape; (<b>B</b>) Semi-arid open woodland; (<b>C</b>) Dry sclerophyll forest with natural and human controlled clearing; (<b>D</b>) Dry sclerophyll forest fragmented by cleared agricultural land. Example locations and FPC colour bar are shown in <a href="#remotesensing-08-00515-f006" class="html-fig">Figure 6</a>.</p> Full article ">
12604 KiB  
Article
A Comparison of Machine Learning Algorithms for Mapping of Complex Surface-Mined and Agricultural Landscapes Using ZiYuan-3 Stereo Satellite Imagery
by Xianju Li, Weitao Chen, Xinwen Cheng and Lizhe Wang
Remote Sens. 2016, 8(6), 514; https://doi.org/10.3390/rs8060514 - 18 Jun 2016
Cited by 110 | Viewed by 10659
Abstract
Land cover mapping (LCM) in complex surface-mined and agricultural landscapes could contribute greatly to regulating mine exploitation and protecting mine geo-environments. However, there are some special and spectrally similar land covers in these landscapes which increase the difficulty in LCM when employing high [...] Read more.
Land cover mapping (LCM) in complex surface-mined and agricultural landscapes could contribute greatly to regulating mine exploitation and protecting mine geo-environments. However, there are some special and spectrally similar land covers in these landscapes which increase the difficulty in LCM when employing high spatial resolution images. There is currently no research on these mixed complex landscapes. The present study focused on LCM in such a mixed complex landscape located in Wuhan City, China. A procedure combining ZiYuan-3 (ZY-3) stereo satellite imagery, the feature selection (FS) method, and machine learning algorithms (MLAs) (random forest, RF; support vector machine, SVM; artificial neural network, ANN) was proposed and first examined for both LCM of surface-mined and agricultural landscapes (MSMAL) and classification of surface-mined land (CSML), respectively. The mean and standard deviation filters of spectral bands and topographic features derived from ZY-3 stereo images were newly introduced. Comparisons of three MLAs, including their sensitivities to FS and whether FS resulted in significant influences, were conducted for the first time in the present study. The following conclusions are drawn. Textures were of little use, and the novel features contributed to improve classification accuracy. Regarding the influence of FS: FS substantially reduced feature set (by 68% for MSMAL and 87% for CSML), and often improved classification accuracies (with an average value of 4.48% for MSMAL using three MLAs, and 11.39% for CSML using RF and SVM); FS showed statistically significant improvements except for ANN-based MSMAL; SVM was most sensitive to FS, followed by ANN and RF. Regarding comparisons of MLAs: for MSMAL based on feature subset, RF achieved the greatest overall accuracy of 77.57%, followed by SVM and ANN; for CSML, SVM had the highest accuracies (87.34%), followed by RF and ANN; based on the feature subsets, significant differences were observed for MSMAL and CSML using any pair of MLAs. In general, the proposed approach can contribute to LCM in complex surface-mined and agricultural landscapes. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Location of study area and field survey samples, and ZiYuan-3 fused true color image (R—Red, G—Green, B—Blue). Jing-Zhu expressway: connecting Beijing and Zhuhai; G107: national highway 107 of China; Hu-Rong expressway: connecting Shanghai and Chengdu; Jing-Guang railway: connecting Beijing and Guangzhou; Wu-Xian inter-city railway: connecting Wuhan city and Xianning of Hubei Province, China; Wu-Guang high-speed railway: connecting Wuhan and Guangzhou.</p> Full article ">Figure 2
<p>Flowchart of methods used in this study. ZY-3: ZiYuan-3; NL: nadir-looking; FL: front looking; BL: backward looking; PAN: panchromatic; MS: multispectral; DTM: digital terrain models; VI: vegetation index; PCs: principal components; GLP filters: the Gaussian low-pass filter features; Mean filters: the mean filter features; StDev filters: the standard deviation filter features; MSMAL: mapping of surface-mined and agricultural landscapes (<span class="html-italic">i.e.</span>, the first-level land covers with gray shades); CSML: classification of surface-mined land (<span class="html-italic">i.e.</span>, the second-level land covers with black shades); RF: random forest; SVM: support vector machine; ANN: artificial neural network.</p> Full article ">Figure 3
<p>Results for the mapping of surface-mined and agricultural landscapes derived from the feature subset-based random forest, support vector machine, and artificial neural network models (top to bottom). Black and white rectangles represent areas with misclassifications.</p> Full article ">Figure 4
<p>Overlay display of results for the classification of surface-mined land derived from the feature subset-based random forest, support vector machine, and artificial neural network models, from top to bottom, on the ZiYuan-3 fused true color image (R—Red, G—Green, B—Blue) scaled to fit the surface-mined land. The yellow numbers 1–12 represent 12 mines.</p> Full article ">Figure 5
<p>Location of test samples for the mapping of surface-mined and agricultural landscapes, and red band of ZiYuan-3 fused image.</p> Full article ">
3712 KiB  
Article
Object-Based Greenhouse Mapping Using Very High Resolution Satellite Data and Landsat 8 Time Series
by Manuel A. Aguilar, Abderrahim Nemmaoui, Antonio Novelli, Fernando J. Aguilar and Andrés García Lorca
Remote Sens. 2016, 8(6), 513; https://doi.org/10.3390/rs8060513 - 18 Jun 2016
Cited by 74 | Viewed by 11854
Abstract
Greenhouse mapping through remote sensing has received extensive attention over the last decades. In this article, the innovative goal relies on mapping greenhouses through the combined use of very high resolution satellite data (WorldView-2) and Landsat 8 Operational Land Imager (OLI) time series [...] Read more.
Greenhouse mapping through remote sensing has received extensive attention over the last decades. In this article, the innovative goal relies on mapping greenhouses through the combined use of very high resolution satellite data (WorldView-2) and Landsat 8 Operational Land Imager (OLI) time series within a context of an object-based image analysis (OBIA) and decision tree classification. Thus, WorldView-2 was mainly used to segment the study area focusing on individual greenhouses. Basic spectral information, spectral and vegetation indices, textural features, seasonal statistics and a spectral metric (Moment Distance Index, MDI) derived from Landsat 8 time series and/or WorldView-2 imagery were computed on previously segmented image objects. In order to test its temporal stability, the same approach was applied for two different years, 2014 and 2015. In both years, MDI was pointed out as the most important feature to detect greenhouses. Moreover, the threshold value of this spectral metric turned to be extremely stable for both Landsat 8 and WorldView-2 imagery. A simple decision tree always using the same threshold values for features from Landsat 8 time series and WorldView-2 was finally proposed. Overall accuracies of 93.0% and 93.3% and kappa coefficients of 0.856 and 0.861 were attained for 2014 and 2015 datasets, respectively. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Location of the study area on a RGB Landsat 8 image path 200 and row 34, covering the “Poniente” region. Coordinate system: ETRS89 UTM Zone 30°N.</p> Full article ">Figure 2
<p>Manually digitized ground truths in ETRS89 UTM coordinate system: (<b>a</b>) Ground Truth for September 2013; and (<b>b</b>) Ground Truth for July 2015.</p> Full article ">Figure 3
<p>Accuracy assessment based on the 1500 segments belonging to the validation set for each considered year and strategy: (<b>a</b>) Overall Accuracy (OA); (<b>b</b>) Kappa coefficient (k); (<b>c</b>) Producer’s accuracy for the Greenhouse class (PA_GH); and (<b>d</b>) User’s accuracy for Greenhouse class (UA_GH).</p> Full article ">Figure 3 Cont.
<p>Accuracy assessment based on the 1500 segments belonging to the validation set for each considered year and strategy: (<b>a</b>) Overall Accuracy (OA); (<b>b</b>) Kappa coefficient (k); (<b>c</b>) Producer’s accuracy for the Greenhouse class (PA_GH); and (<b>d</b>) User’s accuracy for Greenhouse class (UA_GH).</p> Full article ">Figure 4
<p>Threshold Model proposed for classifying greenhouse (GH) based on temporally stable features.</p> Full article ">Figure 5
<p>Overall accuracy achieved for each year through object-based (3000 segments and presenting confidence intervals), pixel-based (ground truth manually digitized) and Threshold Model using features from Landsat 8 and WV2 (TM L8 + WV2).</p> Full article ">Figure 6
<p>Pixel-based accuracy assessment process attained through the threshold model for a detailed area of 950 m × 900 m for each year: (<b>a</b>) segmentation on WV2 MS 2013 orthoimage. B letters indicate white buildings; (<b>b</b>) segmentation on WV2 MS 2015 orthoimage. B letters indicate white buildings; (<b>c</b>) manually digitized ground truth for 2013 (used for 2014 TS) with Greenhouses in red and Non-Greenhouses in green; (<b>d</b>) manually digitized ground truth for 2015 with Greenhouses in red and Non-Greenhouses in green; (<b>e</b>) classification results using the TM approach for 2013–2014 with errors in brown; and (<b>f</b>) classification results using the TM approach for 2015 with errors in brown.</p> Full article ">
2513 KiB  
Article
Mangroves at Their Limits: Detection and Area Estimation of Mangroves along the Sahara Desert Coast
by Viviana Otero, Katrien Quisthoudt, Nico Koedam and Farid Dahdouh-Guebas
Remote Sens. 2016, 8(6), 512; https://doi.org/10.3390/rs8060512 - 18 Jun 2016
Cited by 16 | Viewed by 7152
Abstract
The northernmost and most arid mangrove ecosystem of West Africa is found in Mauritania, in the Parc National du Banc d’Arguin (PNBA). The existing global and regional maps of Mauritania’s mangroves have little detail, and available estimates of the mangrove area differ among [...] Read more.
The northernmost and most arid mangrove ecosystem of West Africa is found in Mauritania, in the Parc National du Banc d’Arguin (PNBA). The existing global and regional maps of Mauritania’s mangroves have little detail, and available estimates of the mangrove area differ among studies. We assessed the use of automated Remote Sensing classification techniques to calculate the extent and map the distribution of the mangrove patches located at Cap Timiris, PNBA, using QuickBird and GeoEye imagery. It was possible to detect the northernmost contiguous mangrove patches of West Africa with an accuracy of 87% ± 2% using the Maximum Likelihood algorithm. The main source of error was the low spectral difference between mangroves and other types of terrestrial vegetation, which resulted in an erroneous classification between these two types of land cover. The most reliable estimate for the mangrove area obtained in this study was 19.48 ± 5.54 ha in 2011. Moreover, we present a special validation procedure that enables a detailed and reliable validation of the land cover maps. Full article
(This article belongs to the Special Issue Remote Sensing in Coastal Environments)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>The study site Cap Timiris at Parc National du Banc d’Arguin (PNBA), Mauritania. The Iwik area and the Parc National du Diawling (PND) are also indicated (see text). Adapted from Google Maps (2016).</p> Full article ">Figure 2
<p>Procedure used to select the pixels for training and validation in the supervised classification based on the cross-validation technique.</p> Full article ">Figure 3
<p>Box-plot of the accuracies for the Maximum Likelihood (ML) and neural network (NN) algorithms implemented for the image from 2004 (<b>A</b>) and 2011 (<b>B</b>). The overall accuracies are named “ML” and “NN”. The specific accuracies are named “ML specific” and “NN specific”. These results were obtained with the five iterations of the cross-validation technique.</p> Full article ">Figure 4
<p>Box-plot showing the quantity, exchange, and shift differences of the image from 2004 (<b>A</b>) and 2011 (<b>B</b>). ML refers to the Maximum Likelihood and NN to the neural network algorithm. These metrics were calculated taking into account the specific accuracy.</p> Full article ">Figure 5
<p>Area (Ha) calculated with Maximum Likelihood Algorithm (ML) and the neural network (NN) for the image from 2004 (<b>A</b>) and in 2011 (<b>B</b>).</p> Full article ">Figure 6
<p>Mangroves at Cap Timiris, since mangroves in PNBA only comprise the mangrove species <span class="html-italic">Avicennia germinans</span>. There is an overestimation of mangroves in the central-east part of the area, mangroves are not found in that location.</p> Full article ">
1670 KiB  
Article
Using the NASA EOS A-Train to Probe the Performance of the NOAA PATMOS-x Cloud Fraction CDR
by Andrew Heidinger, Michael Foster, Denis Botambekov, Michael Hiley, Andi Walther and Yue Li
Remote Sens. 2016, 8(6), 511; https://doi.org/10.3390/rs8060511 - 18 Jun 2016
Cited by 18 | Viewed by 6738
Abstract
An important component of the AVHRR PATMOS-x climate date record (CDR)—or any satellite cloud climatology—is the performance of its cloud detection scheme and the subsequent quality of its cloud fraction CDR. PATMOS-x employs the NOAA Enterprise Cloud Mask for this, which is based [...] Read more.
An important component of the AVHRR PATMOS-x climate date record (CDR)—or any satellite cloud climatology—is the performance of its cloud detection scheme and the subsequent quality of its cloud fraction CDR. PATMOS-x employs the NOAA Enterprise Cloud Mask for this, which is based on a naïve Bayesian approach. The goal of this paper is to generate analysis of the PATMOS-x cloud fraction CDR to facilitate its use in climate studies. Performance of PATMOS-x cloud detection is compared to that of the well-established MYD35 and CALIPSO products from the EOS A-Train. Results show the AVHRR PATMOS-x CDR compares well against CALIPSO with most regions showing proportional correct values of 0.90 without any spatial filtering and 0.95 when a spatial filter is applied. Values are similar for the NASA MODIS MYD35 mask. A direct comparison of PATMOS-x and MYD35 from 2003 to 2014 also shows agreement over most regions in terms of mean cloud amount, inter-annual variability, and linear trends. Regional and seasonal differences are discussed. The analysis demonstrates that PATMOS-x cloud amount uncertainty could effectively screen regions where PATMOS-x differs from MYD35. Full article
(This article belongs to the Special Issue Satellite Climate Data Records and Applications)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Schematic illustration of the co-location process of MYD02SSH (<b>a</b>) and AVHRR/GAC (<b>b</b>) with the NASA Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO)/ Cloud–Aerosol Lidar with Orthogonal Polarization (CALIOP). Larger square grid represents the 1-km (MYD021KM or AVHRR/HRPT) pixels. The red ellipse represent one co-location point. Grey pixels are the MYD02SSH or AVHRR/GAC pixels that contribute to the cloud fraction computation. Orange pixels denote 1-km CALIOP Cloud Layer values that are determined by the co-location process. Yellow pixels are the CALIPSO/CALIOP pixels that contribute to the cloud cloud fraction for this co-location.</p> Full article ">Figure 2
<p>Comparison of AVHRR PATMOS-x and MYD35 for winter 2003–2014 for day and night (all). Panel (<b>a</b>) shows the mean cloud fraction at 2.5° resolution; Panel (<b>b</b>) shows the PATMOS-x–MYD35 difference; Panel (<b>c</b>) shows the PATMOS-x uncertainty from the naïve Bayesian cloud detection scheme; Panel (<b>d</b>) shows the anomaly correlation of PATMOS-x and MYD35. Panel (<b>e</b>) shows the PATMOS-x linear trend, and Panel (<b>f</b>) shows a scatterplot of the PATMOS-x and MYD35 linear trends.</p> Full article ">Figure 3
<p>Comparison of AVHRR PATMOS-x and MYD35 for summer 2003–2014 for day and night (all). Panel (<b>a</b>) shows the mean cloud fraction at 2.5° resolution; Panel (<b>b</b>) shows the PATMOS-x–MYD35 difference; Panel (<b>c</b>) shows the PATMOS-x uncertainty from the naïve Bayesian cloud detection scheme; Panel (<b>d</b>) shows the anomaly correlation of PATMOS-x and MYD35. Panel (<b>e</b>) shows the PATMOS-x linear trend, and Panel (<b>f</b>) shows a scatterplot of the PATMOS-x and MYD35 linear trends.</p> Full article ">Figure 4
<p>Comparison of the PATMOS-x cloud fraction uncertainty climate date record (CDR) to the direct measurements of cloud fraction difference from (<b>a</b>) CALIPSO and (<b>b</b>) MYD35. The 0/100 spatial filter was applied.</p> Full article ">Figure 5
<p>Comparison of AVHRR PATMOS-x and AVHRR PATMOS-x with an increase in assumed climatological cloud fraction (prior cloud probability) by 10% for 2003–2014 for day and night. PATMOS-x refers to the standard AVHRR PATMOS-x run on MYD02SSH and PATMOSx_10 refers to the results from increase in climatological cloud amount. Panel (<b>a</b>) shows the mean cloud fraction at 2.5° resolution; Panel (<b>b</b>) shows the PATMOSx_10 and PATMOS-x difference; Panel (<b>c</b>) shows the PATMOS-x uncertainty from the naïve Bayesian cloud detection scheme; Panel (<b>d</b>) shows the anomaly correlation of PATMOS-x and PATMOXs_10. Panel (<b>e</b>) shows the difference in the linear trends between PATMOSx_10 and PATMOS-x, and Panel (<b>f</b>) shows a scatterplot of the PATMOS-x and PATMOSx_10 linear trends. Global cloud fraction for the two data sets are shown in the legend of Panel (a).</p> Full article ">Figure 6
<p>Comparison of PATMOS-x generated annual mean cloud fractions run on MYD02SSH data (2003–2014). The channel used in PATMOS-x for each sensor is given in <a href="#remotesensing-08-00511-t001" class="html-table">Table 1</a>. The upper left panel (<b>a</b>) is the mean cloud fraction from MODIS; panel (<b>b</b>) is the cloud differences between MODIS and VIIRS; panel (<b>c</b>) is the cloud fraction difference between MODIS and AVHRR/3B; panel (<b>d</b>) is the cloud fraction difference between AVHRR/3B and AVHRR/3A. Global mean values are displayed in the legends on the bottom for each figure.</p> Full article ">
16299 KiB  
Letter
Analysis of Aerosol Radiative Forcing over Beijing under Different Air Quality Conditions Using Ground-Based Sun-Photometers between 2013 and 2015
by Wei Chen, Lei Yan, Nan Ding, Mengdie Xie, Ming Lu, Fan Zhang, Yongxu Duan and Shuo Zong
Remote Sens. 2016, 8(6), 510; https://doi.org/10.3390/rs8060510 - 17 Jun 2016
Cited by 5 | Viewed by 5864
Abstract
Aerosol particles can strongly affect both air quality and the radiation budget of the atmosphere. Above Beijing, the capital city of China, large amounts of aerosols within the atmospheric column have caused the deterioration of local air quality and have influenced radiative forcings [...] Read more.
Aerosol particles can strongly affect both air quality and the radiation budget of the atmosphere. Above Beijing, the capital city of China, large amounts of aerosols within the atmospheric column have caused the deterioration of local air quality and have influenced radiative forcings at both the top and the bottom of the atmosphere (BOA and TOA). Observations of aerosol radiative forcing and its efficiency have been made using two sun-photometers in urban Beijing between 2013 and 2015, and have been analyzed alongside two air quality monitoring stations’ data by dividing air quality conditions into unpolluted, moderately polluted, and heavily polluted days. Daily average PM2.5 concentrations varied greatly in urban Beijing (5.5–485.0 µg/m3) and more than one-third of the analyzed period is classified as being polluted according to the national ambient air quality standards of China. The heavily polluted days had the largest bottom of atmosphere (BOA) and top of atmosphere (TOA) radiative forcings, but the smallest radiative forcing efficiencies, while the unpolluted days showed the opposite characteristics. On heavily polluted days, the averaged BOA aerosol radiative forcing occasionally exceeded −150 W/m2, which represents a value about three-times greater than that for unpolluted days. BOA aerosol radiative forcing was around two-to-three times as large as TOA aerosol radiative forcing under various air quality conditions, although both were mostly negative, suggesting that aerosols had different magnitudes of cooling effects at both the surface and the top of the atmosphere. Unpolluted days had the largest average values of aerosol radiative forcing efficiencies at BOA (and TOA) levels, which exceeded −190 W/m2 (−70 W/m2), compared with the lowest average values in heavily polluted days of around −120 W/m2 (−55 W/m2). These results suggest that the high concentrations of particulate matter pollution in the urban Beijing area had a strong cooling effect at both BOA and TOA levels. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>The distribution of air quality monitoring stations and AERONET stations in Beijing.</p> Full article ">Figure 2
<p>Daily variations of PM<sub>2.5</sub> concentrations (µg/m<sup>3</sup>) for the two selected monitoring stations.</p> Full article ">Figure 3
<p>Monthly average PM<sub>2.5</sub> concentrations (µg/m<sup>3</sup>) for both selected monitoring stations.</p> Full article ">Figure 4
<p>Numbers of polluted and heavily polluted days recorded at both monitoring stations. (<b>a</b>) West Park Official station; (<b>b</b>) Olympic Sports Center station.</p> Full article ">Figure 5
<p>Aerosol radiative forcing at the bottom of the atmosphere (ARF-BOA) for both AERONET stations.</p> Full article ">Figure 6
<p>Aerosol radiative forcing at the top of the atmosphere (ARF-TOA) for both AERONET stations.</p> Full article ">Figure 7
<p>Aerosol radiative forcing efficiencies recorded at the bottom of the atmosphere for both AERONET stations.</p> Full article ">Figure 8
<p>Aerosol radiative forcing efficiencies recorded at the top of the atmosphere for both AERONET stations.</p> Full article ">Figure 9
<p>Aerosol radiative forcings at the bottom and top of the atmosphere under different air quality conditions.</p> Full article ">Figure 10
<p>Aerosol radiative forcing efficiencies at the bottom and top of the atmosphere under different air quality conditions.</p> Full article ">
4407 KiB  
Article
Time Series MODIS and in Situ Data Analysis for Mongolia Drought
by Munkhzul Dorjsuren, Yuei-An Liou and Chi-Han Cheng
Remote Sens. 2016, 8(6), 509; https://doi.org/10.3390/rs8060509 - 16 Jun 2016
Cited by 44 | Viewed by 9411
Abstract
Drought is a period of abnormally dry weather with a serious shortage of water supply. Drought indices can be an advantageous indicator to assess drought for taking further response actions. However, drought indices based on ground meteorological measurements could not completely reveal the [...] Read more.
Drought is a period of abnormally dry weather with a serious shortage of water supply. Drought indices can be an advantageous indicator to assess drought for taking further response actions. However, drought indices based on ground meteorological measurements could not completely reveal the land use effects over a regional scale. On the other hand, the satellite-derived products provide consistent, spatial and temporal comparisons of global signatures for the regional-scale drought events. This research is to investigate the drought signatures over Mongolia by using satellite remote sensing imagery. The evapotranspiration (ET), potential evapotranspiration (PET) and two-band Enhanced Vegetation Index (EVI2) were extracted from MODIS data. Based on the standardized ratio of ET to PET (ET/PET) and EVI2, the Modified Drought Severity Index (MDSI) anomaly during the growing season from May–August for the years 2000–2013 was acquired. Fourteen-year summer monthly data for air temperature, precipitation and soil moisture content of in situ measurements from sixteen meteorological stations for four various land use areas were analyzed. We also calculated the percentage deviation of climatological variables at the sixteen stations to compare to the MDSI anomaly. Both comparisons of satellite-derived and observed anomalies and variations were analyzed by using the existing common statistical methods. The results demonstrated that the air temperature anomaly (T anomaly) and the precipitation anomaly (P anomaly) were negatively (correlation coefficient r = −0.66) and positively (r = 0.81) correlated with the MDSI anomaly, respectively. The MDSI anomaly distributions revealed that the wettest area occupied 57% of the study area in 2003, while the driest (drought) area occurred over 54% of the total area in 2007. The results also showed very similar variations between the MDSI and T anomalies. The highest (wettest) MDSI anomaly indicated the lowest T anomaly, such as in the year 2003, while the lowest (driest) MDSI anomaly had the highest T anomaly in 2007. By comparing the MDSI anomaly and soil moisture content at a 10-cm depth during the study period, it is found that their correlation coefficient is 0.74. Full article
(This article belongs to the Special Issue Earth Observations for a Better Future Earth)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Land use/cover map of Mongolia (41°35′N–52°09′N and 87°44′E–119°56′E).</p> Full article ">Figure 2
<p>Time series of the climate data for air temperature (°C: red open squares), precipitation (mm: blue bars) and soil moisture content (%: light red open triangles) averages over the sixteen stations during the growing season of 2000–2013.</p> Full article ">Figure 3
<p>(<b>a</b>–<b>n</b>) MDSI anomaly distributions during the growing season for the years 2000–2013: The MDSI anomaly was computed as the standardized <span class="html-italic">Z</span> terms represented in Equation (6).</p> Full article ">Figure 4
<p>Frequency distribution of the MDSI anomaly with five categories of the total area during the growing season of 2000–2013: (i) severely wet: &gt;2 (green bars); (ii) moderately wet: 0.5 to 1.99 (light green bars); (iii) normal: 0.49 to −0.49 (white bars); (iv) moderate drought: −0.5 to −1.99 (light yellow bars); and (v) severe drought: &lt;−2 (orange bars).</p> Full article ">Figure 5
<p>Time series for the MDSI anomaly with different climatological variables over the sixteen stations during 2000–2013: (<b>a</b>) anomalies of MDSI (green open circles) and <span class="html-italic">T</span> (red opened squares); and (<b>b</b>) anomalies of MDSI (green open circles) and <span class="html-italic">P</span> (blue bars). The horizontal lines show that the MDSI anomaly value is −0.5 or less for drought conditions, whereas horizontal dashed lines show the percentage deviation values of <span class="html-italic">T</span> (0.0 or above) and <span class="html-italic">P</span> (0.0 or less) for drought condition.</p> Full article ">Figure 6
<p>Time series for the MDSI anomaly (green open circles) with soil moisture content (%: light red open triangles) over the sixteen stations in 2000–2013: the horizontal line shows that the MDSI anomaly value is −0.5 or less for drought condition.</p> Full article ">Figure 7
<p>(<b>a</b>–<b>d</b>) Temporal variations of the MDSI anomaly among different land use areas for 2000–2013: (<b>a</b>) forest steppe; (<b>b</b>) steppe; (<b>c</b>) high mountains; and (<b>d</b>) desert steppe. The MDSI anomaly symbols correspond to the average of four consecutive months’ (the growing season) value calculated in the pixel region at each station. The vertical axes depict the standardized MDSI anomaly (green open circles) of the pixel region. Precipitation (mm: blue bars) is accumulated during the growing season, while soil moisture content (%: light red open triangles) is acquired during the growing season periods to match the MDSI anomaly value in the same periods. The horizontal line shows that the MDSI anomaly value −0.5 or less for drought condition.</p> Full article ">
158 KiB  
Editorial
Preface: Remote Sensing of Biodiversity
by Susan L. Ustin
Remote Sens. 2016, 8(6), 508; https://doi.org/10.3390/rs8060508 - 16 Jun 2016
Cited by 2 | Viewed by 4576
Abstract
Since the 1992 Earth Summit in Rio de Janeiro, the importance of biological diversity insupporting and maintaining ecosystem functions and processes has become increasingly understood [1]. Full article
(This article belongs to the Special Issue Remote Sensing of Biodiversity)
12992 KiB  
Article
Transformation Model with Constraints for High-Accuracy of 2D-3D Building Registration in Aerial Imagery
by Guoqing Zhou, Qingli Luo, Wenhan Xie, Tao Yue, Jingjin Huang and Yuzhong Shen
Remote Sens. 2016, 8(6), 507; https://doi.org/10.3390/rs8060507 - 16 Jun 2016
Cited by 4 | Viewed by 5510
Abstract
This paper proposes a novel rigorous transformation model for 2D-3D registration to address the difficult problem of obtaining a sufficient number of well-distributed ground control points (GCPs) in urban areas with tall buildings. The proposed model applies two types of geometric constraints, co-planarity [...] Read more.
This paper proposes a novel rigorous transformation model for 2D-3D registration to address the difficult problem of obtaining a sufficient number of well-distributed ground control points (GCPs) in urban areas with tall buildings. The proposed model applies two types of geometric constraints, co-planarity and perpendicularity, to the conventional photogrammetric collinearity model. Both types of geometric information are directly obtained from geometric building structures, with which the geometric constraints are automatically created and combined into the conventional transformation model. A test field located in downtown Denver, Colorado, is used to evaluate the accuracy and reliability of the proposed method. The comparison analysis of the accuracy achieved by the proposed method and the conventional method is conducted. Experimental results demonstrated that: (1) the theoretical accuracy of the solved registration parameters can reach 0.47 pixels, whereas the other methods reach only 1.23 and 1.09 pixels; (2) the RMS values of 2D-3D registration achieved by the proposed model are only two pixels along the x and y directions, much smaller than the RMS values of the conventional model, which are approximately 10 pixels along the x and y directions. These results demonstrate that the proposed method is able to significantly improve the accuracy of 2D-3D registration with much fewer GCPs in urban areas with tall buildings. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Error of the 2D-3D registration caused by both various and dissymmetric image distortions, displacement and a rigid 3D object model.</p> Full article ">Figure 2
<p>The geometry for the coplanar condition.</p> Full article ">Figure 3
<p>The geometry of the perpendicular condition.</p> Full article ">Figure 4
<p>Discussion of the relationship between the straight line and the coplanar constraint and the relationship between the perpendicular constraint and the angle-unpreserved perspective projection.</p> Full article ">Figure 5
<p>Six aerial images from two strips in the study area, City of Denver, Colorado [<a href="#B6-remotesensing-08-00507" class="html-bibr">6</a>].</p> Full article ">Figure 6
<p>(<b>a</b>) 2D brightness is applied to represent the digital building model (DBM); and (<b>b</b>) constructive solid geometry (CSG) is used to represent the DBM [<a href="#B6-remotesensing-08-00507" class="html-bibr">6</a>].</p> Full article ">Figure 7
<p>(<b>a</b>) The corners of buildings in the 3D model, which are taken as “GCPs” and (<b>b</b>) the corresponding 2D image coordinates in the 2D image plane.</p> Full article ">Figure 8
<p>The two types of geometric constraints: (<b>a</b>) the linear features in the building model; and (<b>b</b>) the corresponding linear features in the aerial image.</p> Full article ">Figure 9
<p>The experimental results of 2D-3D registration using the proposed method. (<b>a</b>) The 2D-3D registration results with the proposed method; (<b>b</b>) the sub-area of (<b>a</b>) framed by the black rectangle; (<b>c</b>) the sub-area of (<b>b</b>), which illustrates the detail 2D-3D registration results of a single high-rise building.</p> Full article ">Figure 10
<p>Visual comparison of 2D-3D registration using the two models: (<b>a<sub>1</sub></b>–<b>e<sub>1</sub></b>) are the 2D-3D registration results created by the proposed model, in which 32 points plus the two types of constraints are employed; (<b>a<sub>2</sub></b>–<b>e<sub>2</sub></b>) and (<b>a<sub>3</sub></b>–<b>e<sub>3</sub></b>) are the 2D-3D registration results created by conventional models, in which 211 “GCPs” and 12 “GCPs” are employed, respectively.</p> Full article ">
2864 KiB  
Article
Learning a Transferable Change Rule from a Recurrent Neural Network for Land Cover Change Detection
by Haobo Lyu, Hui Lu and Lichao Mou
Remote Sens. 2016, 8(6), 506; https://doi.org/10.3390/rs8060506 - 16 Jun 2016
Cited by 280 | Viewed by 17126
Abstract
When exploited in remote sensing analysis, a reliable change rule with transfer ability can detect changes accurately and be applied widely. However, in practice, the complexity of land cover changes makes it difficult to use only one change rule or change feature learned [...] Read more.
When exploited in remote sensing analysis, a reliable change rule with transfer ability can detect changes accurately and be applied widely. However, in practice, the complexity of land cover changes makes it difficult to use only one change rule or change feature learned from a given multi-temporal dataset to detect any other new target images without applying other learning processes. In this study, we consider the design of an efficient change rule having transferability to detect both binary and multi-class changes. The proposed method relies on an improved Long Short-Term Memory (LSTM) model to acquire and record the change information of long-term sequence remote sensing data. In particular, a core memory cell is utilized to learn the change rule from the information concerning binary changes or multi-class changes. Three gates are utilized to control the input, output and update of the LSTM model for optimization. In addition, the learned rule can be applied to detect changes and transfer the change rule from one learned image to another new target multi-temporal image. In this study, binary experiments, transfer experiments and multi-class change experiments are exploited to demonstrate the superiority of our method. Three contributions of this work can be summarized as follows: (1) the proposed method can learn an effective change rule to provide reliable change information for multi-temporal images; (2) the learned change rule has good transferability for detecting changes in new target images without any extra learning process, and the new target images should have a multi-spectral distribution similar to that of the training images; and (3) to the authors’ best knowledge, this is the first time that deep learning in recurrent neural networks is exploited for change detection. In addition, under the framework of the proposed method, changes can be detected under both binary detection and multi-class change detection. Full article
(This article belongs to the Special Issue Monitoring of Land Changes)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>The pseudocolor images of Taizhou with RGB 432, acquired in (<b>a</b>) March 2000 and (<b>b</b>) February 2003. The labeled ground truth of the Taizhou images: (<b>c</b>) binary ground-truth, where unchanged areas are shown in gray, changed areas are shown in white and black indicates an unlabeled region not used for testing; (<b>d</b>) ground-truth of multi-class changes, where unchanged areas are shown in red, changed areas of city expansion are shown in green, changed soil areas are shown in orange, changed water areas are shown in blue and gray represents unlabeled regions (more details can be found in <a href="#sec4dot3-remotesensing-08-00506" class="html-sec">Section 4.3</a>).</p> Full article ">Figure 2
<p>The pseudocolor images in Kunshan with RGB 432, acquired in (<b>a</b>) March 2000 and (<b>b</b>) February 2003. The labeled ground truth of the Kunshan images: (<b>c</b>) binary ground-truth, where unchanged areas are shown in gray, changed areas are shown in white and black represents unlabeled regions not used for testing; (<b>d</b>) ground-truth of multi-class changes, where unchanged areas are shown in red, changed areas of city expansion are shown in blue, changed farmland areas are shown in green and gray represents unlabeled regions (more details can be found in <a href="#sec4dot3-remotesensing-08-00506" class="html-sec">Section 4.3</a>).</p> Full article ">Figure 3
<p>The images in Yancheng with RGB 432, acquired in (<b>a</b>) May 2006 and (<b>b</b>) April 2007; (<b>c</b>) The labeled binary ground-truth, where unchanged areas are shown in black and changed areas are shown in white.</p> Full article ">Figure 4
<p>The framework overview of the proposed REFEREE change detection model.</p> Full article ">Figure 5
<p>The structure of the LSTM unit in our REFEREEmodel, in which gray arrows indicate directed connection, where the information will flow along this direction, and a gray dotted arrow indicates a peephole connection. Additional details are available in the text.</p> Full article ">Figure 6
<p>An unrolled recurrent neural network.</p> Full article ">Figure 7
<p>The thematic maps obtained from REFEREE with the Taizhou images: (<b>a</b>) binary change map of the whole images, where black indicates unchanged regions and white indicates changed regions; (<b>b</b>) confidence level map of the whole Taizhou image, where the color bar is described in the text; (<b>c</b>) binary change map of the labeled samples, where white indicates changed regions, gray indicates unchanged regions and black indicates unlabeled regions; (<b>d</b>) confidence level map of the labeled samples.</p> Full article ">Figure 8
<p>The thematic maps obtained using REFEREE for the Kunshan images: (<b>a</b>) binary change map of the whole images, where black indicates non-changes and white indicates changes; (<b>b</b>) confidence level map of the whole Kunshan images, where the color bar is described in the text; (<b>c</b>) binary change map of labeled samples, where white indicates changed regions, gray indicates unchanged regions and black indicates unlabeled regions; (<b>d</b>) confidence level map of the labeled samples.</p> Full article ">Figure 9
<p>The thematic maps obtained using REFEREE for the Yancheng images: (<b>a</b>) binary change map of the whole images, where black indicates non-changes and white indicates changes; (<b>b</b>) confidence level map of the whole Kunshan images, where the color bar is described in the text.</p> Full article ">Figure 10
<p>Transfer results for different range numbers of the Kunshan training samples for testing all of the labeled Taizhou images.</p> Full article ">Figure 11
<p>Transfer results for different range numbers of the Taizhou training samples for testing all of the labeled Kunshan images.</p> Full article ">Figure 12
<p>Transfer results for different range numbers of the Taizhou training samples for testing all of the labeled Yancheng images.</p> Full article ">Figure 13
<p>Multi-class change results for the Taizhou images: (<b>a</b>) confidence level map of the city expansion; (<b>b</b>) confidence level map of the soil change; (<b>c</b>) confidence level map of the water change; (<b>d</b>) confidence level map of an unchanged region; (<b>e</b>) changed map detected by REFEREE; and (<b>f</b>) ground truth, where the unchanged areas are shown in red, the changed region of the city expansion is shown in green, the changed soil region is shown in orange and the changed water areas are shown in blue. Additional details about the color bar can be found in the text.</p> Full article ">Figure 14
<p>Multi-class change results of the Kunshan images: (<b>a</b>) confidence level map of the city expansion; (<b>b</b>) confidence level map of the farmland change; (<b>c</b>) confidence level map of an unchanged region; (<b>d</b>) changed map detected by REFEREE; and (<b>e</b>) ground truth, where the unchanged areas are shown in red, the changed region of the city expansion is shown in blue and the changed farmland region is shown in green. Additional details can be found in the text.</p> Full article ">
3873 KiB  
Article
Estimating Snow Water Equivalent with Backscattering at X and Ku Band Based on Absorption Loss
by Yurong Cui, Chuan Xiong, Juha Lemmetyinen, Jiancheng Shi, Lingmei Jiang, Bin Peng, Huixuan Li, Tianjie Zhao, Dabin Ji and Tongxi Hu
Remote Sens. 2016, 8(6), 505; https://doi.org/10.3390/rs8060505 - 16 Jun 2016
Cited by 45 | Viewed by 7490
Abstract
Snow water equivalent (SWE) is a key parameter in the Earth’s energy budget and water cycle. It has been demonstrated that SWE can be retrieved using active microwave remote sensing from space. This necessitates the development of forward models that are capable of [...] Read more.
Snow water equivalent (SWE) is a key parameter in the Earth’s energy budget and water cycle. It has been demonstrated that SWE can be retrieved using active microwave remote sensing from space. This necessitates the development of forward models that are capable of simulating the interactions of microwaves and the snow medium. Several proposed models have described snow as a collection of sphere- or ellipsoid-shaped ice particles embedded in air, while the microstructure of snow is, in reality, more complex. Natural snow usually forms a sintered structure following mechanical and thermal metamorphism processes. In this research, the bi-continuous vector radiative transfer (bi-continuous-VRT) model, which firstly constructs snow microstructure more similar to real snow and then simulates the snow backscattering signal, is used as the forward model for SWE estimation. Based on this forward model, a parameterization scheme of snow volume backscattering is proposed. A relationship between snow optical thickness and single scattering albedo at X and Ku bands is established by analyzing the database generated from the bi-continuous-VRT model. A cost function with constraints is used to solve effective albedo and optical thickness, while the absorption part of optical thickness is obtained from these two parameters. SWE is estimated after a correction for physical temperature. The estimated SWE is correlated with the measured SWE with an acceptable accuracy. Validation against two-year measurements, using the SnowScat instrument from the Nordic Snow Radar Experiment (NoSREx), shows that the estimated SWE using the presented algorithm has a root mean square error (RMSE) of 16.59 mm for the winter of 2009–2010 and 19.70 mm for the winter of 2010–2011. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Comparison of the GB-SAR observed with the model predicted backscattering on 17 January 2007 (<b>a</b>) at X band and (<b>b</b>) at Ku band; and 5 February 2007 (<b>c</b>) at X band and (<b>d</b>) at Ku band.</p> Full article ">Figure 2
<p>Total backscattering coefficient from dry snow at X (<b>left</b>) and Ku (<b>right</b>) bands with VV polarization and incidence angle of 40° against increase in snow depth. The values of 0.1 mm, 0.3 mm, 0.4 mm and 0.6 mm were applied for the optical grain size <math display="inline"> <semantics> <mrow> <msub> <mi>R</mi> <mi>e</mi> </msub> </mrow> </semantics> </math>. The <span class="html-italic">b</span> parameter was 1.2.</p> Full article ">Figure 3
<p>The attenuation factor for different snow densities for X (<b>left</b>) and Ku (<b>right</b>) bands.</p> Full article ">Figure 4
<p>Comparing the simulated snow volume backscattering of X at VV (<b>a</b>) and VH (<b>b</b>) polarizations; and that of Ku at VV (<b>c</b>) and VH (<b>d</b>) polarizations with the ones calculated by Equations (7)–(10), respectively.</p> Full article ">Figure 5
<p>The relationship between single scattering albedo at X band and that at Ku band (<b>left</b>) and the relationship between the snow optical thickness at X bands and at Ku band (<b>right</b>).</p> Full article ">Figure 6
<p>The contours of cost function with constraints (<b>a</b>) and without constraints (<b>b</b>), using a simulated test case.</p> Full article ">Figure 7
<p>Comparison between the empirically estimated absorption coefficient through Equation (16) and the one simulated by the bi-continuous-VRT model.</p> Full article ">Figure 8
<p>The relationship between temperature and SWE at X band under the fixed absorption part of optical thickness 0.0057 at X band.</p> Full article ">Figure 9
<p>The time series of observed SWE (<b>black</b>) and estimated SWE (<b>red</b>) during: 27 December 2009–19 March 2010 (<b>a</b>); and 29 October 2010–4 April 2011 (<b>b</b>).</p> Full article ">Figure 10
<p>Retrieval using six different settings for the albedo reference values and variance values. Retrieved SWE (<b>red</b> dots) in comparison to measured SWE (<b>yellow</b> dots) with ±30 mm standard error (<b>black</b>) during: 27 December 2009–19 March 2010 (<b>a</b>); and 29 October 2010–4 April 2011 (<b>b</b>).</p> Full article ">
6051 KiB  
Article
Estimation of Daily Solar Radiation Budget at Kilometer Resolution over the Tibetan Plateau by Integrating MODIS Data Products and a DEM
by Laure Roupioz, Li Jia, Françoise Nerry and Massimo Menenti
Remote Sens. 2016, 8(6), 504; https://doi.org/10.3390/rs8060504 - 16 Jun 2016
Cited by 25 | Viewed by 7773
Abstract
Considering large and complex areas like the Tibetan Plateau, an analysis of the spatial distribution of the solar radiative budget over time not only requires the use of satellite remote sensing data, but also of an algorithm that accounts for strong variations of [...] Read more.
Considering large and complex areas like the Tibetan Plateau, an analysis of the spatial distribution of the solar radiative budget over time not only requires the use of satellite remote sensing data, but also of an algorithm that accounts for strong variations of topography. Therefore, this research aims at developing a method to produce time series of solar radiative fluxes at high temporal and spatial resolution based on observed surface and atmosphere properties and topography. The objective is to account for the heterogeneity of the land surface using multiple land surface and atmospheric MODIS data products combined with a digital elevation model to produce estimations daily at the kilometric level. The developed approach led to the production of a three-year time series (2008–2010) of daily solar radiation budget at one kilometer spatial resolution across the Tibetan Plateau. The validation showed that the main improvement from the proposed method is a higher spatial and temporal resolution as compared to existing products. However, even if the solar radiation estimates are satisfying on clear sky conditions, the algorithm is less reliable under cloudy sky condition and the albedo product used here has a too coarse temporal resolution and is not accurate enough over rugged terrain. Full article
(This article belongs to the Special Issue Land Surface Processes and Interactions—From HCMM to Sentinel Missions and beyond)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>The Tibetan Plateau and surroundings, with the study area extent delimitation (red line), and location of the ground stations (circled red crosses) within the study area.</p> Full article ">Figure 2
<p>MODIS daily products gap filling procedure.</p> Full article ">Figure 3
<p>Computation of the instantaneous solar radiation budget for all skies using MODIS products and a DEM.</p> Full article ">Figure 4
<p>Surface irradiance computed using: (<b>a</b>) a constant atmospheric transmission factor or (<b>b</b>) the proposed methodology and the solar geometry expressed as: (<b>c</b>) the solar zenith angle or (<b>d</b>) the sun incident angle.</p> Full article ">Figure 5
<p>Time series of instantaneous solar incident (yellow) and reflected (purple) radiations for the pixels in which the ground stations used for the validation are located: BJ, Linzhi, NamCo, and Qomolangma (Qomo).</p> Full article ">Figure 6
<p>Time series of surface albedo derived from MODIS (MOD43B3) over the four ground stations used for the validation: BJ (blue), Linzhi (red), NamCo (green), and Qomolangma (yellow).</p> Full article ">Figure 7
<p>Solar radiative fluxes validation for instantaneous incident (orange) and reflected (purple) fluxes as well as daily net solar radiation (green). The validation is provided for all skies (<b>left</b>) or clear sky (<b>right</b>) conditions, at BJ, Qomo, Linzhi, and NamCo stations for the entire time series.</p> Full article ">Figure 8
<p>Time series of estimated (yellow) and measured (green) instantaneous solar incident radiations for each of the pixel used for the comparison with the ground stations of BJ, NamCo, and Qomolangma (Qomo).</p> Full article ">Figure 9
<p>Comparison of the differences in W·m<sup>−2</sup> between estimated and measured instantaneous incoming fluxes at BJ, NamCo, and Qomolangma (Qomo) according to the input missing during the computation.</p> Full article ">Figure 10
<p>Comparison between TOA reflected radiance estimated using MODIS surface reflectance and TOA reflected radiance measured by MODIS sensor: (<b>a</b>) mean TOA reflected radiance estimated (purple) and measured (green) for the entire study area; (<b>b</b>) the regression between the both mean TOA reflected radiance.</p> Full article ">Figure 11
<p>The four ground stations location on the Tibetan Plateau (yellow pin) and the footprint of the pixel used for the validation (red lines). BJ (31°22’N, 91°53’E), Linzhi (29°45’N, 94°44’E), NamCo (30°46’N, 90°59’E), Qomo (28°21’N, 86°56’E).</p> Full article ">
1073 KiB  
Erratum
Erratum: Cavender-Bares, J.; Meireles, J.E.; Couture, J.; Kaproth, M.A.; Kingdon, C.C; Singh, A; Serbin, S.P.; Center, A; Zuniga, E; Pilz, G; Townsend, P.A. Associations of Leaf Spectra with Genetic and Phylogenetic Variation in Oaks: Prospects for Remote Detection of Biodiversity. Remote Sens. 2016, 8, 221
by Remote Sensing Editorial Office
Remote Sens. 2016, 8(6), 475; https://doi.org/10.3390/rs8060475 - 16 Jun 2016
Viewed by 3575
Abstract
The authors would like to correct the abstract and Figures 3 and 4 of this article [1] as follows:[...] Full article
►▼ Show Figures

Figure 3

Figure 3
<p>(<b>A</b>) Bivariate plot of population means (±1 SE) from a principal coordinates (PCO) analysis PCO using leaf spectra from five populations of <span class="html-italic">Quercus oleoides</span> (BZ = Belize, MX = Mexico, HN = Honduras, CR-SE = Costa Rica, Santa Elena, and CR-RI = Costa Rica, Rincon), showing the first and second axes of variation. The first two components were significantly differentiated by population (see Table S2); (<b>B</b>) PCO scores showing population means (±1 SE) for the first and fourth principal components as black circles with 95% CI. These were the first two components that had significant phylogenetic signal (see Table S2); (<b>C</b>) PCO scores for the first and fourth principal components shown for the four higher order clades (live oaks, <span class="html-italic">Virentes</span> (<span class="html-italic">V</span>, green symbol); white oaks, section <span class="html-italic">Quercus</span> (<span class="html-italic">Q</span>, blue symbol), red oaks, section <span class="html-italic">Lobatae</span> (<span class="html-italic">L</span>, red symbol), and golden cup oaks, section <span class="html-italic">Protobalanus</span> (<span class="html-italic">P</span>, gold symbol); (<b>D</b>) PCO scores for leaf type (evergreen (E), deciduous (D) or brevi-deciduous (BD)), showing the first and fourth principle components, the first two components that were significantly differentiated by leaf type (see Table S3).</p> Full article ">Figure 4
<p>(<b>A</b>) Molecular phylogeny of 28 oak species showing principal coordinate scores to the right of each species. Leaf habit is indicated by letter codes and coloring of species names, as follows: D = deciduous (red), E = evergreen (dark green), and BD = brevi-deciduous or semi-evergreen (light green). The four recognized higher level clades are indicated with colored circles as follows: red, red oaks (section <span class="html-italic">Lobatae</span>); blue, white oaks (section <span class="html-italic">Quercus</span>); green, live oaks (<span class="html-italic">Virentes</span>); and yellow, golden cup oaks (<span class="html-italic">Protobalanus</span>). Species values for the first and fourth principal coordinate (PCO) axes are shown to the right: (<b>B</b>) PCO1 and PCO4. Positive PCO axis values are shown in dark gray, negative in light gray. Distributions of observed values of Blomberg’s K statistic (red dashed lines) are shown relative to a Brownian motion (BM) model of evolution (dark gray bars) and relative to a white noise model in which phylogenetic relationships are completely randomized (light gray bars) for (<b>C</b>) PCO1 (<b>left</b>) and PCO4 (<b>right</b>) species scores. Observed K values for the PCO1 and PCO4 scores are consistent with a Brownian motion model of evolution and show higher phylogenetic conservatism than expected based on random relationships.</p> Full article ">
1714 KiB  
Article
Improving Streamflow Prediction Using Remotely-Sensed Soil Moisture and Snow Depth
by Haishen Lü, Wade T. Crow, Yonghua Zhu, Fen Ouyang and Jianbin Su
Remote Sens. 2016, 8(6), 503; https://doi.org/10.3390/rs8060503 - 15 Jun 2016
Cited by 14 | Viewed by 5933
Abstract
The monitoring of both cold and warm season hydrologic processes in headwater watersheds is critical for accurate water resource monitoring in many alpine regions. This work presents a new method that explores the simultaneous use of remotely sensed surface soil moisture (SM) and [...] Read more.
The monitoring of both cold and warm season hydrologic processes in headwater watersheds is critical for accurate water resource monitoring in many alpine regions. This work presents a new method that explores the simultaneous use of remotely sensed surface soil moisture (SM) and snow depth (SD) retrievals to improve hydrological modeling in such areas. In particular, remotely sensed SM and SD retrievals are applied to filter errors present in both solid and liquid phase precipitation accumulation products acquired from satellite remote sensing. Simultaneously, SM and SD retrievals are also used to correct antecedent SM and SD states within a hydrological model. In synthetic data assimilation experiments, results suggest that the simultaneous correction of both precipitation forcing and SM/SD antecedent conditions is more efficient at improving streamflow simulation than data assimilation techniques which focus solely on the constraint of antecedent SM or SD conditions. In a real assimilation case, results demonstrate the potential benefits of remotely sensed SM and SD retrievals for improving the representation of hydrological processes in a headwater basin. In particular, it is demonstrated that dual precipitation/state correction represents an efficient strategy for improving the simulation of cold-region hydrological processes. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>The flowchart for data assimilation cases C1–C10.</p> Full article ">Figure 2
<p>Location of the Tuotuo River Basin and Ganzi Basin in Western China.</p> Full article ">Figure 3
<p>HBV model simulated and observed average monthly streamflow within designated calibration (2007–2010) and validation (2011–2013) periods at the outlet of the (<b>A</b>) Tuotuo River Basin and the (<b>B</b>) Ganzi Basin.</p> Full article ">Figure 4
<p>The RMSE of corrected (<b>A</b>) SMART- and (<b>B</b>) SDART-corrected precipitation for various levels of (synthetically-introduced) precipitation error.</p> Full article ">Figure 5
<p>Daily streamflow flow RMSE and correlation (<span class="html-italic">R</span>) results for synthetic data assimilation experiments based on the application of the Open Loop (OL) HBV model (with no DA) and the 10 DA cases (C1 to C10) outlined in <a href="#sec2dot6-remotesensing-08-00503" class="html-sec">Section 2.6</a> and <a href="#remotesensing-08-00503-f001" class="html-fig">Figure 1</a>.</p> Full article ">Figure 6
<p>Daily observed streamflow at the outlet of the Tuotuo River Basins for both: (<b>A</b>) Open Loop HBV simulated results obtained without data assimilation and (<b>B</b>) case C10 data assimilation results between 1 May 2007 and 30 September 2010.</p> Full article ">Figure 7
<p>Daily observed streamflow at the outlet of the Ganzi Basin for both: (<b>A</b>) Open Loop HBV simulated results obtained without data assimilation and (<b>B</b>) case C10 data assimilation results between 1 January 2010 and 21 December 2013.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-90
Previous Issue
Next Issue
Back to TopTop