Search

[Display omitted] The increasing availability of remote sensing imagery led to the emergence of numerous burned area mapping products and methodologies, thus, the necessity of spatial and temporally validating them. The main objective of this paper is to conduct a global analysis and comparison of MODIS and VIIRS active fire dates with dating provided by moderate resolution burned area mapping products. As reference burned area maps, we used the Burned Area Reference Database (BARD) to analyze and compare the performances of MODIS and VIIRS active fires. For 316 path-rows, we calculated the temporal gap between the dates of BARD pixels and those of the five nearest active fires. Olson terrestrial biomes were taken into account in the analysis of results. The agreement rates between the dates of burned areas and of active fires varied based on their geographical distance, fire size and biome. The highest agreement rates were observed for smaller distances (<500 m) and larger fires (>100 ha). Boreal and Mediterranean biomes exhibited the highest values of date percent agreement between burned area pixels and active fires, while the lowest values were found in biomes encompassed by grasslands and savannas. [ABSTRACT FROM AUTHOR]

Accurate evapotranspiration (ET) estimations based on surface energy balance from remote sensing require information in the thermal infrared (TIR) domain, normally provided with an insufficient spatial resolution. In order to estimate ET in heterogeneous agricultural areas, we inspect in this paper the use of disaggregation techniques applied to two different sensors, such as MODIS (daily revisit cycle and 1 km spatial resolution in the TIR domain) and Spot 5 (5 days revisit cycle and 10 m spatial resolution in the VNIR bands but no TIR band). Spot 5 images were used as a proxy for upcoming Sentinel-2. The Simplified Two-Source Energy Balance (STSEB) model was used for the estimation of ET. Since no Sentinel-2 images were available yet, images from the Spot 5 Take 5 experiment were used for testing this approach. Results assessment was conducted at two different levels: field scale (using ground data), and scene scale (using Landsat 7-ETM + images as a reference). Validation of both disaggregated land surface temperature (LST) and derived surface energy fluxes was performed. Mean absolute deviations of ~ 2 °C in disaggregated LST were observed at both field and scene scales. At field scale, relative errors of 22% and 19% were obtained for ET at instantaneous and daily scales, respectively. At scene scale, the four components of the surface energy balance equation were obtained with relative errors of 3, 14, 11 and 8% for net radiation, evapotranspiration, sensible heat flux and soil heat flux, respectively, compared to Landsat. The results obtained were compared to the use of the MODIS LST at its original resolution (1 km), which was used also to obtain the surface energy fluxes. As the surface heterogeneity increases the errors in both MODIS LST and ET become more and more significant, compared to the use of the disaggregated images. Although reference images at 10 m spatial resolution were not available at this stage for a more robust comparison, this paper shows the potential of the use of disaggregated LST to estimate ET at 10 m spatial resolution, which is especially attractive in highly heterogeneous areas. [ABSTRACT FROM AUTHOR]

Earthquake is one of the most devastating natural disasters that its prediction has not materialized comprehensive. Remote sensing data can be used to access information which is closely related to an earthquake. The unusual variations of lithosphere, atmosphere and ionosphere parameters before the main earthquakes are considered as earthquake precursors. To date the different precursors have been proposed. This paper examines one of the parameters which can be derived from satellite imagery. The mentioned parameter is Aerosol Optical Depth (AOD) that this article reviews its relationship with earthquake. Aerosol parameter can be achieved through various methods such as AERONET ground stations or using satellite images via algorithms such as the DDV (Dark Dense Vegetation), Deep Blue Algorithm and SYNTAM (SYNergy of Terra and Aqua Modis). In this paper, by analyzing AOD’s time series (derived from MODIS sensor on the TERRA platform) for 16 major earthquakes, seismic anomalies were observed before and after earthquakes. Before large earthquakes, rate of AOD increases due to the pre-seismic changes before the strong earthquake, which produces gaseous molecules and therefore AOD increases. Also because of aftershocks after the earthquake there is a significant change in AOD due to gaseous molecules and dust. These behaviors suggest that there is a close relationship between earthquakes and the unusual AOD variations. Therefore the unusual AOD variations around the time of earthquakes can be introduced as an earthquake precursor. [ABSTRACT FROM AUTHOR]

For regions where limited in-situ observations are available for use as a reference, such as in China, the understanding of how changes in the Moderate Resolution Imaging Spectroradiometer (MODIS) aerosol retrieval algorithm impact the data quality is limited. In this paper, the MODIS-Aqua Aerosol Optical Thickness (AOT) Collection 5 (C5) and 6 (C6) Dark-Target (DT) products are evaluated by comparison with similar products from the China Aerosol Remote Sensing NETwork (CARSNET) and the Aerosol Robotic Network (AERONET) in China for the year 2008. The validation data are analyzed for key factors affecting the results of the MODIS-DT aerosol retrieval algorithm, including surface type, aerosol loading and aerosol absorption. The impacts of the change from the C5 to the C6 versions of the MODIS AOT retrieval algorithm are discussed in detail. The main findings are: (1) The agreement from retrieval aspects between both MODIS C5/C6 and CARSNET-AERONET for most AOT values over distinct regions indicates that the conclusions obtained from C5 hold for C6; (2) MODIS-DT has an extended coverage (more retrievals) with good quality due to the change of upstream data (calibration, cloud masking, etc); (3) About 64% of the MODISDT C6 AOT data over China fall within the (±0.15 AOT ± 0.05) error envelope; (4) positive AOT differences between C6 and C5 over relatively dark surfaces and negative differences over relatively bright surfaces due to the change of surface parameterization have been observed; (5) the AOT difference between MODIS and CARSNET-AERONET relating to AOT loading is not significant, especially for large AOT; (6) MODIS-DT performs well for absorbing aerosol over relatively bright surfaces. • MODIS/Aqua AOT C5/C6 DT products are evaluated to understand the effects of different parameters on algorithm performance. • Both MODIS C5 and C6 DT show good quality over China under different surface and atmospheric conditions. • MODIS C6 DT aerosol product has an extended coverage with good quality over China. [ABSTRACT FROM AUTHOR]

Abstract: Traditional methods of recording fire burned areas and fire severity involve expensive and time-consuming field surveys. Available remote sensing technologies may allow us to develop standardized burn-severity maps for evaluating fire effects and addressing post fire management activities. This paper focuses on multiscale characterization of fire severity using multisensor satellite data. To this aim, both MODIS (Moderate Resolution Imaging Spectroradiometer) and ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) data have been processed using geo-statistic analyses to capture pattern features of burned areas. Even if in last decades different authors tried to integrate geo-statistics and remote sensing image processing, methods used since now are only variograms, semivariograms and kriging. In this paper, we propose an approach based on the use of spatial indicators of global and local autocorrelation. Spatial autocorrelation statistics, such as Moran''s I and Getis–Ord Local Gi index, were used to measure and analyze dependency degree among spectral features of burned areas. This approach enables the characterization of pattern features of a burned area and improves the estimation of fire severity. [Copyright &y& Elsevier]

The current paper presents the development and testing of a multi-step methodology which integrates remotely sensed and ancillary data to estimate olive (Olea europaea L.) fruit yield in Tuscany (Central Italy). The processing of very high resolution (Ikonos) and high resolution (Landsat ETM+) images provides a map of olive tree canopy cover fraction for all Tuscany olive yards, which is used to extract olive tree NDVI values from MODIS imagery. The combination of these values with standard meteorological data within a modified parametric model (C-Fix) enables the prediction of daily olive tree gross primary production (GPP) for ten years (2000–2009). These GPP estimates are then joint to the respiration estimates of a bio-geochemical model (BIOME-BGC) to simulate olive tree net primary production (NPP). The NPP accumulated over proper periods of the ten growing seasons is finally converted into olive fruit yield, whose accuracy is assessed through comparison with provincial statistics. The methodology is only partly capable of capturing spatial and temporal olive fruit yield variability at province level, but can accurately reproduce inter-year yield variation over the entire region. The paper concludes with a discussion of the results achieved and with considerations on the research prospects. [Copyright &y& Elsevier]

• The geography of SST time series summarizes the complex processes at play in Eastern Boundary Currents. • A Dynamic Time Warping-based method applied to dense MODIS time series was used to partition the ocean. • The partition was compared with a frontal activity climatology created for the same area. • Method provides first reported co-location of dynamic regional boundaries and frontal activity maxima. • The new methodology is scalable and based in freely accessible data available worldwide. Biogeographic partitioning systems provide an important framework under which ocean management practices can be designed and implemented. Sea Surface Temperature (SST) provides important insights into the features and processes governing biogeography and the natural boundaries that separate regions (fronts). In this paper, a new methodology is suggested to segment the ocean at a subset of an Eastern Boundary Current system (Western Iberian Peninsula). To this end, a dense time series of daily MODIS data (Sea Surface Temperature (SST) and Chlorophyll-a concentration, 2003–2014) was analyzed using Dynamic Time Warping (DTW). The resulting dissimilarity matrix was then clustered using UPGMA to generate a set of dynamic regions. To determine whether the boundaries separating regions coincided with areas of maximum frontal activity, a matching time series was produced for the study area. The methodology successfully delivered a set of nested regions, explained by known oceanographic features and separated by natural boundaries collocated with frontal activity maxima (ANOVA, p < 0.01). The frontal activity, which was also found to be spatially organized (Moran's I > 0.82 and z-score > 6.95, for a p < 0.01) was thus correlated with the dissimilarity patterns of SST. Together, they offer a new perspective into the biogeographic constraints influencing an important ecosystem. SST-based regions were more informative (k = 4) but both the Chlorophyll-a and the multivariate (SST and Chlorophyll-a) alternatives reproduced the coastal/oceanic dichotomy of established systems in a dynamic way. The method was also applied to yearly subsets, which generally matched the regions delineated with the full time series while depicting inter-annual variability. The DTW and the frontal activity maps contributed to the definition of a scalable and ecologically-meaningful system, which can be adopted for the monitoring of marine areas using different (annual to decadal) time scales and easily accessible datasets. [ABSTRACT FROM AUTHOR]

Highlights • Ten machine learning methods were tested for retrieval of evapotranspiration from remote sensing. • The best results were obtained with bootstrap aggregation tree, multi-layer neural network, and random kernel methods. • Training with same surface type data did not improve algorithm performance. Abstract Remote sensing retrieval of evapotranspiration (ET), or surface latent heat exchange (LE), is of great utility for many applications. Machine learning (ML) methods have been extensively used in many disciplines, but so far little work has been performed systematically comparing ML methods for ET retrieval. This paper provides an evaluation of ten ML methods for estimating daily ET based on daily Global LAnd Surface Satellite (GLASS) radiation data and high-level Moderate-Resolution Imaging Spectroradiometer (MODIS) data products and ground measured ET data from 184 flux tower sites. Measurements of accuracy (RMSE, R2, and bias) and run time were made for each of ten ML methods with a smaller training data set (n = 7910 data points) and a larger training data set (n= 69,752 data points). Inclusion of more input variables improved algorithm performance but had little effect on run time. The best results were obtained with the larger training data set using the bootstrap aggregation (bagging) regression tree (validation RMSE = 19.91 W/m2) and three hidden layer neural network (validation RMSE = 20.94 W/m2), although the less computationally demanding random kernel (RKS) algorithm also produced good results (validation RMSE = 22.22 W/m2). Comparison of results from sites with different ecosystem types showed the best results for evergreen, shrub, and grassland sites, and the weakest results for wetland sites. Generally, performance was not improved by training with data from only the same ecosystem type. [ABSTRACT FROM AUTHOR]

Abstract Air temperature is an important parameter affecting biosphere processes, air quality, regional climate, and human health, but is difficult to estimate, especially on a nationwide scale. Remote sensing provides high spatial resolution data that influence air temperature and can help overcome this problem in spatially heterogeneous environments. Generally, the simple linear relationships between air temperature, land surface temperature (LST), and auxiliary data, seldom consider the complex spatial variations in air temperature. The random forest (RF) regression model can characterize the complex temperature distribution by establishing the non-linear relationships among them. The main purpose of this paper is to present an RF regression method to estimate monthly average temperatures in China, using MODIS LST, normalized differential vegetation index (NDVI), night-time light, and a digital elevation model (DEM). Results based on cross-validation show that mean absolute error (MAE) and root mean square error (RMSE) at national level in different months ranged from 1.15 to 1.44 °C (mean ± standard deviation: 1.30 ± 0.10 °C) and from 1.57 to 1.99 °C (1.79 ± 0.13 °C), respectively. Potential sources of error and uncertainty were analyzed and investigated. Results show that regression model performs better in eastern (MAE: 0.68 ± 0.79 °C; RMSE: 0.75 ± 0.81 °C) and central China (MAE: 0.68 ± 0.41 °C; RMSE: 0.74 ± 0.41 °C). The model performance was strongly dependent on elevation, slope, land cover, and satellite data quality. In addition, the percentage increase in mean squared error from RF regression model was used to quantify the importance of variables. Positive importance scores indicated an improvement in accuracy with the assistance of NDVI, night-time lights, and the DEM. LST, and more auxiliary data sources, when integrated into an RF regression model, may be ways in which to improve the modeling of air temperature variations in the future. [ABSTRACT FROM AUTHOR]

Abstract This paper introduces a modular processing chain to derive global high-resolution maps of leaf traits. In particular, we present global maps at 500 m resolution of specific leaf area, leaf dry matter content, leaf nitrogen and phosphorus content per dry mass, and leaf nitrogen/phosphorus ratio. The processing chain exploits machine learning techniques along with optical remote sensing data (MODIS/Landsat) and climate data for gap filling and up-scaling of in-situ measured leaf traits. The chain first uses random forests regression with surrogates to fill gaps in the database (> 45 % of missing entries) and maximizes the global representativeness of the trait dataset. Plant species are then aggregated to Plant Functional Types (PFTs). Next, the spatial abundance of PFTs at MODIS resolution (500 m) is calculated using Landsat data (30 m). Based on these PFT abundances, representative trait values are calculated for MODIS pixels with nearby trait data. Finally, different regression algorithms are applied to globally predict trait estimates from these MODIS pixels using remote sensing and climate data. The methods were compared in terms of precision, robustness and efficiency. The best model (random forests regression) shows good precision (normalized RMSE≤ 20 %) and goodness of fit (averaged Pearson's correlation R = 0.78) in any considered trait. Along with the estimated global maps of leaf traits, we provide associated uncertainty estimates derived from the regression models. The process chain is modular, and can easily accommodate new traits, data streams (traits databases and remote sensing data), and methods. The machine learning techniques applied allow attribution of information gain to data input and thus provide the opportunity to understand trait-environment relationships at the plant and ecosystem scales. The new data products – the gap-filled trait matrix, a global map of PFT abundance per MODIS gridcells and the high-resolution global leaf trait maps – are complementary to existing large-scale observations of the land surface and we therefore anticipate substantial contributions to advances in quantifying, understanding and prediction of the Earth system. Highlights • Presented a modular process chain for plant trait mapping including local effects • High-resolution global maps of leaf traits by fusing measured trait data, LANDSAT and MODIS • Scope for testing and parameterizing trait-enabled Earth System models • Implications for land management and Earth system science applications [ABSTRACT FROM AUTHOR]

Abstract: In this paper we analyze the differences obtained in the atmospheric correction of optical imagery covering bands located in the Visible and Near Infra-Red (VNIR), Short-Wave Infra-Red (SWIR) and Themal-Infrared (TIR) spectral regions when atmospheric profiles extracted from different sources are used. In particular, three sensors were used, Compact High Resolution Imaging Spectrometer (CHRIS), Advanced Spaceborne Thermal Emission and Reflection radiometer (ASTER) and Landsat5 Thematic Mapper (TM), whereas four atmospheric profiles sources were considered: i) local soundings launched near the sensor overpass time, ii) Moderate Resolution Radiometer (MODIS) atmospheric profiles product (MOD07), iii) Atmospheric Correction Parameter Calculator (ACPC) generated by the National Center for Environmental Prediction (NCEP) and iv) Modified Atmospheric Profiles from Reanalysis Information (MAPRI), which includes data from NCEP and National Center of Atmospheric Research (NCAR) Reanalysis project but interpolated to 34 atmospheric levels and resampled to 0.5°×0.5°. MODIS aerosol product (MOD04) was also used to extract Aerosol Optical Thickness (AOT) values at 550nm. Analysis was performed for three test dates (12th July 2003, 18th July 2004 and 13th July 2005) over an agricultural area in Spain. Results showed that air temperature vertical profiles were similar for the four sources, whereas dew point temperature profiles showed significant differences at some particular levels. Atmospheric profiles were used as input to MODTRAN4 radiative transfer code in order to compute atmospheric parameters involved in atmospheric correction, with the aim of retrieving surface reflectances in the case of VNIR and SWIR regions, and Land Surface Temperature (LST) in the case of the TIR region. For the VNIR and SWIR region, significant differences depending on the atmospheric profile used were not found, particularly in the Visible region in which the AOT content is the main parameter involved in the atmospheric correction. In the case of TIR, differences depending on the atmospheric profile used were appreciable, since in this case the main parameter involved in the atmospheric correction is the water vapor content, which depends on the vertical profile. In terms of LST retrieval from ASTER data (2004 test case), all profiles provided satisfactory results compared to the ones obtained when using a local sounding, with errors of 0.3K for ACPC and MAPRI cases and 0.7K for MOD07. When retrieving LST from TM data (2005 test case), errors for MOD07 and MAPRI were 0.6 and 0.9K respectively, whereas ACPC provided an error of 2K. The results presented in this paper show that the different atmospheric profile sources are useful for accurate atmospheric correction when local soundings are not available. In particular, MOD07 product provides atmospheric information at the highest spatial resolution, 5km, although its use is limited from 2000 to present, whereas MAPRI provides historical information from 1970 to present, but at lower spatial resolution. [Copyright &y& Elsevier]

Abstract: The generation of multi-decade long Earth System Data Records (ESDRs) of Leaf Area Index (LAI) and Fraction of Photosynthetically Active Radiation absorbed by vegetation (FPAR) from remote sensing measurements of multiple sensors is key to monitoring long-term changes in vegetation due to natural and anthropogenic influences. Challenges in developing such ESDRs include problems in remote sensing science (modeling of variability in global vegetation, scaling, atmospheric correction) and sensor hardware (differences in spatial resolution, spectral bands, calibration, and information content). In this paper, we develop a physically based approach for deriving LAI and FPAR products from the Advanced Very High Resolution Radiometer (AVHRR) data that are of comparable quality to the Moderate resolution Imaging Spectroradiometer (MODIS) LAI and FPAR products, thus realizing the objective of producing a long (multi-decadal) time series of these products. The approach is based on the radiative transfer theory of canopy spectral invariants which facilitates parameterization of the canopy spectral bidirectional reflectance factor (BRF). The methodology permits decoupling of the structural and radiometric components and obeys the energy conservation law. The approach is applicable to any optical sensor, however, it requires selection of sensor-specific values of configurable parameters, namely, the single scattering albedo and data uncertainty. According to the theory of spectral invariants, the single scattering albedo is a function of the spatial scale, and thus, accounts for the variation in BRF with sensor spatial resolution. Likewise, the single scattering albedo accounts for the variation in spectral BRF with sensor bandwidths. The second adjustable parameter is data uncertainty, which accounts for varying information content of the remote sensing measurements, i.e., Normalized Difference Vegetation Index (NDVI, low information content), vs. spectral BRF (higher information content). Implementation of this approach indicates good consistency in LAI values retrieved from NDVI (AVHRR-mode) and spectral BRF (MODIS-mode). Specific details of the implementation and evaluation of the derived products are detailed in the second part of this two-paper series. [Copyright &y& Elsevier]

Abstract: The significance of crop yield estimation is well known in agricultural management and policy development at regional and national levels. The primary objective of this study was to test the suitability of the method, depending on predicted crop production, to estimate crop yield with a MODIS-NDVI-based model on a regional scale. In this paper, MODIS-NDVI data, with a 250m resolution, was used to estimate the winter wheat (Triticum aestivum L.) yield in one of the main winter-wheat-growing regions. Our study region is located in Jining, Shandong Province. In order to improve the quality of remote sensing data and the accuracy of yield prediction, especially to eliminate the cloud-contaminated data and abnormal data in the MODIS-NDVI series, the Savitzky–Golay filter was applied to smooth the 10-day NDVI data. The spatial accumulation of NDVI at the county level was used to test its relationship with winter wheat production in the study area. A linear regressive relationship between the spatial accumulation of NDVI and the production of winter wheat was established using a stepwise regression method. The average yield was derived from predicted production divided by the growing acreage of winter wheat on a county level. Finally, the results were validated by the ground survey data, and the errors were compared with the errors of agro-climate models. The results showed that the relative errors of the predicted yield using MODIS-NDVI are between −4.62% and 5.40% and that whole RMSE was 214.16kgha−1 lower than the RMSE (233.35kgha−1) of agro-climate models in this study region. A good predicted yield data of winter wheat could be got about 40 days ahead of harvest time, i.e. at the booting-heading stage of winter wheat. The method suggested in this paper was good for predicting regional winter wheat production and yield estimation. [Copyright &y& Elsevier]

Abstract: Metrics that estimate the beginning of the growing season in semi-arid monsoonal ecosystems have become a central part of both crop models that are used to monitor crop production using satellite remote sensing and early warning indicators of possible future reductions in yield due to growing season length. This paper presents a new phenological model which is tuned to the semi-arid, monsoonal ecosystem of the West African Sahel and that uses humidity instead of calendar dates to identify the beginning of the season. We implement this model on vegetation index data from AVHRR, SPOT-Vegetation and MODIS data, and evaluate the results along with a rainfall-based start of season estimate for the region. Contributions of the paper include evaluating existing methods using ground observations of sowing date, comparing the same metric across multiple sensors and to rainfall-based SOS, and the development of a metric based on NDVI and relative humidity. [Copyright &y& Elsevier]

This paper aims to test the ability of the WRF-Chem model to simulate sand storms in Northwest China and improve the simulation results by introducing remote sensing soil moisture data into WRF-Chem. We conducted sensitivity tests, including parameterization scheme tests and soil moisture tests using WRF-Chem. By comparing various combinations of parameterization schemes, which have considerable impacts on sand emissions, the most suitable parameterization scheme for WRF-Chem to simulate sand storms in Northwest China was selected for establishing models in future research. Based on the optimal combination of parameterization schemes, the sensitivity of sand emissions to soil moisture was tested, indicating that soil moisture has a substantial influence on sand emissions and revealing a nonlinear relationship between sand emissions and soil moisture. Although sand simulations are sensitive to soil moisture, the sand emission volume does not always increase with decreasing soil moisture content. Moreover, the soil moisture content simulated by WRF-Chem is quite different from that observed by satellites. Therefore, we selected ten sand storms that occurred in Northwest China in 2009–2018, including weak, moderate and strong sand storms, to explore the responses of sand storms of different intensities to soil moisture. Some experiments were conducted in two scenarios: scenario A was simulated only by WRF-Chem, and scenario B was designed by replacing the initial soil moisture field in WRF-Chem with soil moisture satellite data from AMSR2 and AMSR-E. Comparing the simulation results with in-situ ground-based and satellite-based measurements indicates that WRF-Chem can capture strong sand storms and some moderate sand storms, but the ability to simulate weak sand storms needs to be greatly improved. After introducing remote sensing soil moisture data into WRF-Chem, the simulation of sand emissions becomes more accurate, substantially improving the simulation results. Among the ten simulated sand storms, the AOD (aerosol optical depth) and PM10 simulation accuracies are improved for five and seven sand storms, respectively. After this improvement, the AOD correlation coefficient can reach 0.63, and the PM10 correlation coefficient can reach 0.60. • The paper screened out the best parameterization scheme of sand storm simulation that is suitable for northwestern China. • Improved the simulation results of the WRF-Chem model using soil moisture remote sensing data. • WRF-Chem can better capture strong and some moderate sand storms, but need to improve weak sand storm simulation ability. [ABSTRACT FROM AUTHOR]

Land surface temperature (LST) is one of the most important parameters in land surface processes. Although satellite-derived LST can provide valuable information, the value is often limited by cloud contamination. In this paper, a two-step satellite-derived LST reconstruction framework is proposed. First, a multi-temporal reconstruction algorithm is introduced to recover invalid LST values using multiple LST images with reference to corresponding remotely sensed vegetation index. Then, all cloud-contaminated areas are temporally filled with hypothetical clear-sky LST values. Second, a surface energy balance equation-based procedure is used to correct for the filled values. With shortwave irradiation data, the clear-sky LST is corrected to the real LST under cloudy conditions. A series of experiments have been performed to demonstrate the effectiveness of the developed approach. Quantitative evaluation results indicate that the proposed method can recover LST in different surface types with mean average errors in 3–6 K. The experiments also indicate that the time interval between the multi-temporal LST images has a greater impact on the results than the size of the contaminated area. [ABSTRACT FROM AUTHOR]

Correlation analysis has been widely used to validate the accuracy of synthetic medium-resolution data generated by spatial and temporal fusion methods. However, as the temporal resolution of Landsat data is 16 days, this method only can be used for the validation of multi-temporal Landsat data. This means that the fusion accuracy of most images in a synthetic daily time series have not been validated. Furthermore, the fusion accuracy of each image in a synthetic Landsat time series is different, because there is a negative correlation with the time interval length between the fusion date and the base image date. Therefore, there is a need for temporal validation of synthetic daily Landsat time series data. We propose a suitable validation method in this paper. The improved spatial and temporal data fusion approach (ISTDFA) was applied to generate synthetic daily Landsat Normalized Difference Vegetation Index (NDVI) time series, which were then validated for both spatial and temporal dimensions using actual MODIS NDVI time series. For temporal validation, the correlation coefficients ( R ) between the actual and synthetic 500 m NDVI time series were calculated by pixel-by-pixel to generate imagery with an R value for each pixel. For spatial validation, R between MODIS NDVI imagery and synthetic Landsat NDVI imagery was calculated day by day to generate an R time series. This method was tested and validated in two locations (Bole and Luntai) in Xinjiang Province, China. The results show that, for temporal validation, the R values of 86.08% pixels in Bole and 94.71% pixels in Luntai are higher than 0.9, and in spatial validation, R values are higher than 0.8 on most days. Synthetic daily Landsat NDVI data was used to monitor the phenology of vegetation at a spatial resolution of 30 m successfully, while the MODIS product is limited to 500 m. [ABSTRACT FROM AUTHOR]

The major global biogeophysical products obtained from remote sensing are usually composites of 8 or 16 days, and are almost always retrieved from a single sensor. Subsequently, the applications of these products are limited in cases requiring higher temporal frequency. With the increasing number of freely available moderate resolution remote sensing datasets, multi-sensor synergy to increase temporal sampling is warranted. However, radiometric consistency of multi-sensor data is not as good as expected; therefore, first the constituency should be evaluated. A new cross-calibration approach of reflective bands for moderate resolution remotely sensed data is proposed in this paper. The new approach uses a time series of MODIS data from both Terra and Aqua to retrieve both the bidirectional reflectance distribution function (BRDF) and aerosol optical depth (AOD) simultaneously over an invariant desert target. The MODIS retrieved BRDF and AOD are then used to cross-calibrate the medium resolution spectral imager (MERSI) and visible and infra-red radiometer (VIRR) onboard FengYun3-A and B satellites (FY3A/B) and advanced very high resolution radiometer (AVHRR) onboard NOAA-16, 17, and 18. The cross-calibration method is validated in two ways: 1) by comparing with the vicarious calibration; and 2) the top of atmosphere (TOA) reflectance before and after calibration, which show that the calibration error of this new approach is consistent within 5%. Compared with most cross-calibration methods, our new method is generic and better consider on BRDF models; furthermore, the well-built BRDF of the calibration site and simultaneously retrieved AOD make the method fully automatic and cost effective for the calibration of most moderate resolution remote sensors, especially Chinese sensors. [ABSTRACT FROM AUTHOR]

Classification is a fundamental process in remote sensing used to relate pixel values to land cover classes present on the surface. Over large areas land cover classification is challenging particularly due to the cost and difficulty of collecting representative training data that enable classifiers to be consistent and locally reliable. A novel methodology to classify large volume Landsat data using high quality training data derived from the 500 m MODIS land cover product is demonstrated and used to generate a 30 m land cover classification for all of North America between 20°N and 50°N. Publically available 30 m global monthly Web-enabled Landsat Data (GWELD) products generated from every available Landsat 7 ETM + and Landsat 5 TM image for a three year period, that are defined aligned to the MODIS land products and are consistently pre-processed data (cloud-screened, saturation flagged, atmospherically corrected and normalized to nadir BRDF adjusted reflectance), were classified. The MODIS 500 m land cover product was filtered judiciously, using only good quality pixels that did not change land cover class in 2009, 2010 or 2011, followed by automated selection of spatially corresponding 30 m GWELD temporal metric values, to define a large training data set sampled across North America. The training data were sampled so that the class proportions were the same as the North America MODIS land cover product class proportions and corresponded to 1% of the 500 m and < 0.005% of the 30 m pixels. Thirty nine GWELD temporal metrics for every 30 m North America pixel location were classified using (a) a single random forest, and (b) a locally adaptive method with a random forest classifier derived and applied locally and the classification results spatially mosaicked together. The land cover classification results appeared geographically plausible and at synoptic scale were similar to the MODIS land cover product. Detailed visual inspection revealed that the locally adaptive random forest classifications and associated classification confidences were generally more coherent than the single random forest classification results. The level of agreement between the 30 m classifications and the MODIS land cover product derived training data was assessed by bootstrapping the random forest implementation. The locally adaptive random forest classification had higher overall agreement (95.44%, 0.9443 kappa) than the single random forest classification (93.13%, 0.9195 kappa). The paper concludes with a discussion of future research including the potential for automated global land cover classification. [ABSTRACT FROM AUTHOR]

It is well known that remote sensing techniques hold the potential to explore the spatial estimation of air temperature ( T a ) with fine spatial and temporal resolution across the world. However, because of the complex interaction of land-atmosphere system and the contamination of cloud cover, the retrieval of daytime T a exclusively from remote sensing data is still far from straight forward, especially under cloudy sky conditions. In this paper, we presented a simple parameterization scheme of daytime T a under all-weather conditions entirely based on Moderate Resolution Imaging Spectroradiometer (MODIS) products. To evaluate its applicability, the scheme was demonstrated in two regions with totally different geomorphological and climatic conditions, the east part of the Qaidam Basin (EQB) in China and the Southern Great Plains (SGP) in the United States of America. The instantaneous T a under clear sky conditions ( T a , clear ) was determined as the average of near surface air temperature ( T a S ) retrieved from MOD07_L2 product and land surface temperature ( T s ) retrieved from MOD06_L2 product. Then a regression model between T a , clear and T s was established, and the instantaneous T a under cloudy sky conditions ( T a , cl oudy ) was estimated by applying the regression model to T s retrieved under cloudy sky conditions. The results showed that the averaging parameterization scheme has significantly improved the accuracy of T a , clear retrievals with MAE = 1.95 °C, RMSE = 2.50 °C, and B = 0.02 in the EQB, and MAE = 2.02 °C, RMSE = 2.56 °C, and B = 0.01 in the SGP. The T a , cloudy estimates also showed good agreement with T a observations in both regions with a correlation coefficient ( r ) higher than 0.91. The values of RMSE calculated for the EQB and SGP were 3.42 °C and 2.91 °C, respectively. The accuracy of both T a , clear and T a , cloudy estimates has reached a level comparable with other traditional statistical approaches that adopt ancillary T a measurements as training dataset. Therefore, it is feasible to estimate daytime T a under all-weather conditions entirely based on MODIS products. [ABSTRACT FROM AUTHOR]

In this paper, we present a quantitative investigation of the contributions of different aerosols to the aerosol optical depths (AODs) in Taiwan using a global chemical transport model (GEOS-Chem) and remote sensing measurements. The study focus is on the period from June 2012 to October 2013. Five different types of aerosols are investigated: sea salt, dust, sulfate, organic carbon and black carbon. Three of these aerosols, namely sulfate, organic carbon and black carbon, have significant anthropogenic sources. Model simulation results were compared with both ground based sun photometer measurements and MODerate resolution Imaging Spectroradiometer (MODIS) satellite observations. The model data shows good agreement with satellite observations (R = 0.72) and moderate correlation with sun photometer measurements (R = 0.52). Simulation results show the anthropogenic aerosols contribute ∼65% to the total AOD in Taipei, while natural originated aerosols only show a minor impact (∼35%). Among all the aerosols, sulfate is the dominating species, contributing 62.4% to the annual average total AOD. Organic carbon and black carbons respectively contribute 7.3% and 1.5% to the annual averaged total AOD. The annual average contributions of sea salt and dust aerosols to the total AOD are 26.4% and 2.4%, respectively. A sensitivity study was performed to identify the contributions of anthropogenic aerosol sources in each region to the AODs in Taipei. North-East Asia was identified as the major contributing source region of anthropogenic aerosols to Taipei, accounting for more than 50% of total sulfate, 32% of total organic carbon and 51% of total black carbon aerosols. South-East Asia is the second largest contributing source region, contributing 35%, 24% and 34% of total sulfate, organic carbon and black carbon aerosols, respectively. The aerosols from continents other than Asia only show minor impacts to the aerosol load in Taipei. In addition, a case study of a biomass burning episode in South-East Asia during 2013 was investigated. [ABSTRACT FROM AUTHOR]

Remote Sensing observations have enormous advantages in aerosol studies since aerosols' space and time variation. MODIS and CALIOP are two independent instruments with different design principles that provide aerosol optical depth (AOD) retrievals and scan the same points on the Earth's surface within a 2-min interval. Due to predefined aerosol models and fixed vertical profiles in the MODIS algorithm and AOD CALIOP resolution, MODIS and CALIOP cannot give suitable spatial-temporal coverage in related studies. This paper proposes a method based on Bayesian networks to retrieve the AODs by the synergy of CALIOP and MODIS in two vertical layers, 1.5 and 3 km. We applied the Bayesian network for three days in 2018 over the Persian Gulf. The overall analyses reveal that estimated AOD by the seasonal networks correlates with obtained retrieval CALIOP AODs. The correlation values, 0.94 and 0.84, are obtained for the first layer in the summer and winter. These values for the second layer are 0.88 and 0.82. The observed differences in the estimated AOD with the actual measured AOD values and the overall correlation results demonstrate that the proposed networks are sufficient to provide accurate AODs in the two, 1.5, and 3 km vertical layers. According to the experimental results, the layering MAIAC AOD product becomes more suitable for monitoring and studying aerosol phenomena by applying the proposed networks. [Display omitted] •AOD is sensitive to the aerosol vertical profile. •Layering MODIS AOD product using Bayesian networks is a way to afford the challenge of the low temporal-spatial resolution retrieved CALIOP AOD. •Assuming univariate and multi-variate Gaussian distribution for nodes can improve the accuracy of the estimated AODs. •Layering MODIS AOD can give specific knowledge when the increase in aerosol and aerosol transport globally and atmosphere behavior related studies. [ABSTRACT FROM AUTHOR]

Fine particulate matter (PM 2.5 ) can penetrate deeper into the respiratory systems and cause various health problems. In this paper satellite-derived PM 2.5 concentrations, which provide better spatial coverage in the form of satellite-imageries are used to analyse the spatial and temporal distributions of PM 2.5 in Saudi Arabia. PM 2.5 concentrations (μg/m 3 ) are estimated using the relationship between Aerosol Optical Depth (AOD) and PM 2.5 concentrations from satellite images, such as those of the Moderate Resolution Imaging Spectroradiometer (MODIS). PM 2.5 concentrations varied both temporally and spatially and there was a negative south to north trend in PM 2.5 levels. Dammam showed the highest whereas Tabuk showed the lowest PM 2.5 concentrations. Temporally all cities demonstrated a positive trend, except At-Taif and Madinah. The positive trend was significant only in Dammam, Hofuf, Khobar, and Nijran. In most of the cities due to lack of data, ground level PM 2.5 concentrations could not be compared with satellite-derived data, except in Makkah, where a comparison is made between observed and satellite-derived data for years 2001–2007. Both sets of data in Makkah showed positive trends, however satellite-derived concentrations were lower roughly by a factor of 2.5. Remote sensing successfully supplements the ground level air quality monitoring programme and helps better understand the spatial variability of atmospheric pollutants, especially on a large scale, such as regional or global scale. Further comparison between observed and satellite-derived data is required over a larger spatial and temporal resolution. [ABSTRACT FROM AUTHOR]

Although agricultural burning is banned in Russia, it is still a widespread practice. Accurately monitoring cropland burned area is an important task as these estimates are used in the calculation of cropland burning emissions, which are ultimately utilized in policy making decisions. In this paper we developed an independent estimate of cropland burning in Russia through assessing the capabilities of global burned area products (MCD45A1; Roy et al., 2008 and MCD64A1; Giglio et al., 2009) and our own custom Moderate Resolution Imaging Spectroradiometer (MODIS) based Cropland Regional Area Burned (CRAB) product. An archive of cropland field state samples was generated from very high resolution (< 5 m) imagery allowing us a unique perspective into the challenges of mapping cropland burned area through detailed analysis of the Russian agricultural practices. Our analysis showed all three burned area products were unable to map approximately 95% of burn validation samples, demonstrating that the current coarse resolution (defined here as ≥ 500 m) satellite capabilities are not adequate for accurately mapping burned area associated with agricultural fires. The transient nature of cropland burns and the climatological conditions of the region require multiple subsequent daily overpasses; however, the small spatial scale of the burns requires moderate (defined here as 10–50 m) spatial resolutions. Future opportunities to accurately map cropland burned area may arise with the Sentinel-2 and Landsat constellation missions. [ABSTRACT FROM AUTHOR]

Sargassum washing ashore on the beaches of the Caribbean Islands since 2011 has caused problems for the local environments, tourism, and economies. Although preliminary results of Sargassum distributions in the nearby oceans have been obtained using measurements from the Medium Resolution Imaging Spectrometer (MERIS), MERIS stopped functioning in 2012, and detecting and quantifying Sargassum distributions still face technical challenges due to ambiguous pixels from clouds, cloud shadows, cloud adjacency effect, and large-scale image gradient. In this paper, a novel approach is developed to detect Sargassum presence and to quantify Sargassum coverage using the Moderate Resolution Imaging Spectroradiometer (MODIS) alternative floating algae index (AFAI), which examines the red-edge reflectance of floating vegetation. This approach includes three basic steps: 1) classification of Sargassum -containing pixels through correction of large-scale gradient, masking clouds and cloud shadows, and removal of ambiguous pixels; 2) linear unmixing of Sargassum -containing pixels; and, 3) statistics of Sargassum area coverage in pre-defined grids at monthly, seasonal, and annual intervals. In the absence of direct field measurements to validate the results, limited observations from the Hyperspectral Imager for the Coastal Ocean (HICO) measurements and numerous local reports support the conclusion that the elevated AFAI signals are due to the presence of Sargassum instead of other floating materials, and various sensitivity analyses are used to quantify the uncertainties in the derived Sargassum area coverage. The approach was applied to MODIS observations between 2000 and 2015 over the Central West Atlantic (CWA) region (0–22°N, 63–38°W) to derive the spatial and temporal distribution patterns as well as the total area coverage of Sargassum . Results indicate that the first widespread Sargassum distribution event occurred in 2011, consistent with previous MERIS findings. Since 2011, only 2013 showed a minimal Sargassum coverage similar to the period of 2000 to 2010; all other years showed significantly more coverage. More alarmingly, the summer months of 2015 showed mean coverage of > 2000 km 2 , or about 4 times of the summer 2011 coverage and 20 times of the summer 2000 to 2010 coverage. Analysis of several environmental variables provided some hints on the reasons causing the inter-annual changes after 2010, yet further multi-disciplinary research (including in situ measurements) is required to understand such changes and long-term trends in Sargassum coverage. [ABSTRACT FROM AUTHOR]

• Two models are combined to estimate the net carbon fluxes of managed pastures. • The first is an NDVI-driven parametric model, the second a biogeochemical model. • The simulation takes into account the effects of mowing and grazing. • The carbon flux estimates are assessed versus eddy covariance observations. • The approach is able to simulate the carbon balance of managed mountain pastures. The current paper presents a methodology which can simulate the net carbon fluxes of semi-natural grasslands based on the combination of a parametric and a bio-geochemical model. The first model (modified C-Fix) estimates grassland gross primary production (GPP) by driving the light use efficiency approach with remotely sensed normalized difference vegetation index (NDVI) data. The latter model (BIOME-BGC) simulates all main ecosystem processes based on meteorological and ancillary data. The outputs of the two models are finally combined and corrected for the effects of human-induced biomass changes (i.e. mowing, grazing, etc.). The modelling strategy, fed with ancillary and MODIS NDVI data, is tested in two semi-natural grasslands, the first (Amplero) on the Apennines and the second (Monte Bondone) on the Alps (Central and Northern Italy, respectively). The two sites, which show diversified eco-climatic conditions and are managed differently (i.e. Amplero by spring mowing plus cattle grazing and Monte Bondone by summer mowing), were selected for the availability of eddy covariance observations of GPP and net ecosystem production (NEP) collected during several years (from 2003 to 2012). These observations were used to assess the estimates of the modelling strategy, obtaining promising results for both study sites. The greatest estimation errors are found in the Alpine site and are mostly induced by the inaccurate detection of local grassland NDVI values, which is due to the insufficient spatial and temporal resolutions of the used MODIS NDVI imagery. The results of the study highlight the importance of information on the local management practices applied, which is decisive for quantifying medium-term changes in net carbon fluxes. [ABSTRACT FROM AUTHOR]

Snow cover is a key hydrological variable, critical to understanding water cycles and informing management decisions around resource extraction and recreational activities. Remote sensing open-access data and cloud-based computing platforms are two innovative tools for snow cover estimation. In this paper, we present SnowWarp, a processing framework that uses Google Earth Engine and the R programming languages to combine Landsat 30 m with MODIS 500 m satellite imagery and produce daily-30-m spatial resolution snow cover data anywhere globally. SnowWarp was applied in an alpine catchment in Northern Italy from 2000-2019 and validated using hydrometeorological datasets. Strong correlations between snow cover and ground data were found with correlations in terms of R up to −0.84 for temperature, −0.17 for precipitation, 0.74 for snow depth, and −0.43 for streamflow. The SnowWarp tool is an open-source framework enabling users to map fine spatial and temporal dynamics of snow cover to the ecosystem and hydrological monitoring. • The new open-source SnowWarp tool, utilizing Google Earth Engine and R is presented. • Daily snow cover data at 30 m spatial resolution can be calculated globally. • SnowWarp exploits MODIS and Landsat remote sensing data and implements a dynamic time warping algorithm. • SnowWarp products were generated and tested in an Alpine area in Northern Italy. • Snow cover dynamics were related to different hydro-meteorological variables. [ABSTRACT FROM AUTHOR]

The growing popularity of digital-repeat photography in field research is seeing traditional field efforts being assisted and even replaced by low-cost cameras. The efficiency of using cameras is obvious, but there is an assumption that they capture the same information as observations made by humans. This paper aims to determine the level of agreement between these two methods of interpreting understory vegetation phenology. We compared daily phenological observations made by low-cost cameras with those made by personnel during field visits every 10 days. Phenophases were defined as the non-spectral, physical developmental stages of Canadian buffaloberry ( Shepherdia canadensis ) and alpine sweetvetch ( Hedysarum alpinum ). The relationship between observation methods was quantified using a weighted kappa statistic at three spatial scales ranging from individual plants to areas up to 6 ha. Agreement between the camera observations and those made by field personnel was nearly perfect (Kappa > 0.9) for both the vegetative and reproductive phenology of both study species at all spatial scales. The level of agreement was found to be more variable early in the season when plant growth is more rapid. Overall there was a slight bias in the image interpretations to underestimate the rate of development. Time-lapse photography was found to be an analogous replacement for field visits; however, some plant species are more suitable for observation by camera than others. Spatially, it was determined that observations of a single plant are all that is required to capture the phenology of the surrounding region in excess of 6 ha. This analysis was carried out over a single growing season in the in the Rocky Mountains of western Alberta, Canada. [ABSTRACT FROM AUTHOR]

Subjective visual interpretation of MODIS Quick-Look imagery has proven useful in characterising dust source geomorphology and dust event frequency over diverse areas, but is limited by the small temporal sampling window of polar orbiting platforms, and obscuration by cloud. In this paper we seek to quantify the inevitable under-reporting of dust events in Quick-Look imagery, via a three-year case study over the Lake Eyre Basin (LEB), the largest dust source in Australia and the southern hemisphere. Dust event identification from MODIS imagery is compared with the level of dust mobilization inferred from measurements of near-surface aerosol made by a continuously-operating integrating nephelometer, located at the AeroSpan/AERONET station at Tinga Tingana in the Strzelecki Desert, adjacent to major dust sources in the LEB. The analysis indicates a major upward revision in the overall dust event frequency by ~ 72%, including a doubling of the incidence of major dust outbreaks. The intensity of missed events increases toward late afternoon and evening, although without a net day-night bias. These factors suggest the potential of new geostationary sensors and the analysis of night-time polar orbiting imagery in improving the reporting of dust mobilization over active dust-source regions. For the LEB, the corrections are crucial in understanding not only the seasonal distribution of dust mobilization, but its interannual variability and climatology. [ABSTRACT FROM AUTHOR]

The accelerated warming of the Arctic climate may alter the local and regional surface energy balances, for which changing land surface temperatures (LSTs) are a key indicator. Modeling current and anticipated changes in the surface energy balance requires an understanding of the spatio-temporal interactions between LSTs and land cover, both of which can be monitored globally by measurements from space. This paper investigates the accuracy of the MODIS LST/Emissivity Daily L3 Global 1 km V005 product and its spatio-temporal sensitivity to land surface properties in a Canadian High Arctic permafrost landscape. The land cover ranged from fully vegetated wet sedge tundra to barren rock. MODIS LSTs were compared with in situ radiometer measurements from wet tundra areas collected over a 2-year period from July 2008 to July 2010 including both summer and winter conditions. The accuracy of the MODIS LSTs was − 1.1 °C with a root mean square error of 3.9 °C over the entire observation period. Agreement was lowest during the freeze-back periods where MODIS LST showed a cold bias likely due to the overrepresentation of clear-sky conditions. A multi-year analysis of LST spatial anomalies, i.e., the difference between MODIS LSTs and the MODIS LST regional mean, revealed a robust spatio-temporal pattern. Highest variability in LST anomalies was found during freeze-up and thaw periods as well as for open water surface in early summer due to the presence or absence of snow or ice. The summer anomaly pattern was similar for all three years despite strong differences in precipitation, air temperature and net radiation. Summer periods with regional mean LSTs above 5.0 °C showed the greatest spatial diversity with four distinct 2.0 °C classes. Summer anomalies ranged from − 4.5 °C to 2.6 °C with an average standard deviation of 1.8 °C. Dry ridge areas heated up the most, while wetland areas and dry areas of sparsely vegetated bedrock with a high albedo remained coolest. The observed summer LST anomalies can be used as a baseline against which to evaluate both past and future changes in land surface properties that relate to the surface energy balance. Summer anomaly classes mainly reflected a combination of albedo and surface wetness. The potential to use this tool to monitor surface drying and wetting in the Arctic should therefore be further explored. A multi-sensor approach combining thermal satellite measurements with optical and radar imagery promises to be an effective tool for a dynamic, process-based ecosystem monitoring scheme. [ABSTRACT FROM AUTHOR]

In this paper, satellite remote sensing was used as the input parameter of the GRAMI rice model to evaluate its applicability for simulating paddy rice crop condition and yield assessment at the field scale. Especially, the world’s first Geostationary Ocean Color Imager (GOCI), which provides better temporal resolution than does MODIS, was applied to evaluate the estimation of intuitive paddy rice growth and development and to examine the feasibility for vegetation index profiles of the GRAMI rice model. Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data at 500-m resolution were used as reference data to validate the quality of the crop growth and development data derived from GOCI. Field measurements of paddy rice at Chonnam National University, Gwangju, South Korea, were performed to determine initial parameters of the GRAMI rice model, which is used to optimize biophysical processes in the soil–crop–atmosphere system. For angular-dependent vegetation products, daily rolling time series of vegetation indices of GOCI and MODIS were estimated using semi-empirical BRDF modeling based on 16-day composite procedures. The observed temporal variation in GOCI vegetation indices (VIs) based on BAR (bidirectional reflectance distribution function adjusted reflectance) showed a similar growing pattern to the simulated VIs of the crop model, but MODIS showed a difference between measured and simulated VIs during the cloudy monsoon season. The rice yields predicted by integrating satellite data and the GRAMI rice model were compared with field measurements and showed reasonable agreement with reference to paddy rice productivity in the study area. [ABSTRACT FROM AUTHOR]

Timely and accurate monitoring of the onset and evolution of drought in China are important to reduce losses from drought. The Integrated Surface Drought Index (ISDI) which originally established in mideast China shows a large potential for real-time regional drought monitoring. However, ISDI is still at the developmental stage, and the applicability of the index requires further examination especially for China with vast area, climatic conditions, complex topography, and land cover. Furthermore, ISDI model depends on the historical training data corresponding to the study area. ISDI application in China must be remodeled using the historical training data over China. In this paper, we remodeled ISDI over China based on previous work and evaluated its capability for near real-time drought monitoring. Using the Palmer Drought Severity Index (PDSI) as a dependent variable, ISDI integrates climate-based drought indices, satellite-based Vegetation Index (VI) and land surface temperature (LST) with other biophysical and elevation data to produce a 1-km regional drought condition map at 16-day intervals. Strong relationships were determined between the calculated ISDI and PDSI for spring, summer and autumn, and all of the correlation coefficients exceeded 0.8. The initial ISDI results demonstrated a good performance for monitoring droughts in southwestern China from 2009 to 2010, high temperatures and droughts in southern China in 2013, and floods in northeastern China in 2013. The higher spatial resolution and near real-time capability of ISDI can provide important inputs for drought management and mitigation in China. [ABSTRACT FROM AUTHOR]

Fire products derived from coarse (500 m to 1 km) spatial resolution satellite data have become an important source of information for the fire science and applications communities. There is however a demand for moderate (30 m) spatial resolution burned area data sets, systematically generated at regional to global scale, that to date has been only partially met. This paper presents a methodology to fuse multi-temporal Landsat Enhanced Thematic Mapper plus (ETM +) data with Moderate Resolution Imaging Spectroradiometer (MODIS) active fire detections to map systematically burned areas at 30 m resolution. A multistage mapping approach is used with an initial per-pixel change detection based on spectral-rule based pre-classification of Landsat 30 m time series to identify candidate burned areas. The candidate burned area objects are then either retained or discarded by comparison with contemporaneous MODIS active fire detections. The methodology is demonstrated for 1.9 million km 2 over the Western United States using all the Landsat 7 ETM + data and Terra MODIS active fire detections acquired in 2002. Systematic evaluation conducted by per-pixel and per-burned area object comparison with the burned area perimeters provided by the USGS Monitoring Trends in Burn Severity (MTBS) project is presented, and shows a good agreement in the identification of burning patterns but a likely underestimation of the total area burned. Future research to refine and further test the methodology is discussed. [ABSTRACT FROM AUTHOR]

There are extensive reports of ionospheric disturbances before the great 2008 Wenchuan earthquake, which are possibly explained by seismogenic electric field hypotheses linked with the aerosols injected in atmosphere. This paper attempts to investigate the possible change of atmospheric aerosol optical depth (AOD) associated with this earthquake by using MODIS data from both Terra and Aqua satellites. The result shows a clear enhancement of AOD along the Longmenshan faults 7 days before the quake, which is 1 day and 4 days earlier than the reported negative and positive ionospheric disturbances, respectively, and is 1 day earlier than or quasi-synchronism with other reported atmospheric anomalies including air temperature, outgoing longwave radiation and relative humidity. Particularly, the spatial distribution of AOD enhancement is very local and it is correlated well with the active faults and surface ruptures. We suggest that this unique enhancement could be associated with the Lithosphere–Atmosphere–Ionosphere coupling process during the preparation of the Wenchuan earthquake. [ABSTRACT FROM AUTHOR]

The Tibetan Plateau (TP), as a remote and sparsely populated area, is regularly exposed to polluted air masses sourcing from surrounding regions. Atmospheric circulation, as the major driving force generating long-range transport processes of air pollutants, contributes to high-pollution episodes on the TP. Therefore, using reanalysis data from the European Centre for Medium-Range Weather Forecasts for the 2000–2019 period, this paper first classified atmospheric circulation patterns over the study area into nine types (type 1 – type 9). Among them, circulation types 1, 2, 6, and 8 mainly occurred in spring and winter, while circulation types 3, 4, 5, 7, and 9 primarily occurred in summer and autumn. Second, ground-based and satellite remote sensing data were combined to investigate the impact of atmospheric circulation patterns on the properties of aerosols over Central-West Asia and their surrounding areas. We detailed how the atmospheric circulation patterns impacted the aerosol optical depth, Ångström exponent, and aerosol types at different Aerosol Robotic Network sites in the study area. The results obtained from ground-based data were further verified by those from satellite remote sensing data. Third, backward trajectories and the corresponding potential source contribution function based on the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model were used to explore the impact of atmospheric circulation patterns on regional transport pathways of aerosols. It was found that under circulation types 1, 2, 6, and 8, few HYSPLIT trajectories were sourced from the south direction, while under circulation types 3, 4, 5, 7, and 9, the trajectories originating from the south increased significantly, which could be attributed to the summer monsoon. • Impact of circulation patterns on the properties of aerosols was poorly studied. • Effect of circulation patterns on AOD, AE, and aerosol type were revealed. • Impact of circulation patterns on transport pathways of aerosols were explored. • This study enriches theoretical connotation of transport of aerosols in the TP. [ABSTRACT FROM AUTHOR]

Government and NGO funded conservation programs are being implemented in developing countries with the potential benefit of reduced sediment inflow into fresh water lakes. However, these claims are difficult to verify due to limited historical sediment concentration data in lakes and rivers. Remote sensing can potentially aid in monitoring sediment concentration. With almost daily availability over the past ten years and consistent atmospheric correction applied to the images, Moderate Resolution Imaging Spectroradiometer (MODIS) 250 meter images are potential resources capable of monitoring future concentrations and reconstructing historical sediment concentration records. In this paper, site-specific relationships are developed between reflectance in near-infrared (NIR) images and three factors: total suspended solids (TSS), turbidity and Secchi depth for Lake Tana near the mouth of the Gumara River. The first two sampling campaigns on November 27, 2010 and May 13, 2011 are used in calibration. Reflectance in the NIR varies linearly with turbidity ( R 2 = 0.89) and TSS ( R 2 = 0.95). Secchi depth fit best to an exponential relation with R 2 of 0.74. The relationships are validated using a third sample set collected on November 7, 2011 with RMSE of 11 Nephelometric Turbidity Units (NTU) for Turbidity, 16.5 mg l −1 for TSS and 0.12 meters for Secchi depth. The MAE was 10% for TSS, 14% for turbidity and 0.1% for Secchi depth. Using the relationship for TSS, a 10-year time series of sediment concentration in Lake Tana near the Gumara River was plotted. It was found that after the severe drought of 2002 and 2003 the concentration in the lake increased significantly. The results showed that MODIS images are potential cost effective tools to monitor suspended sediment concentration and obtain a past history of concentration for evaluating the effect of best management practices. [ABSTRACT FROM AUTHOR]

Abstract: Ocean color (OC) remote sensing has entered a new phase with the successful deployment of the Visible Infrared Imager Radiometer Suite (VIIRS) sensor aboard the Suomi National Polar-orbiting Partnership (SNPP) satellite. The representativeness and accuracy of the VIIRS geophysical products need to be assessed before a wide use of these data by the scientific community. As an integral part of the VIIRS sensor calibration and validation efforts, our group has been continuously monitoring the validity of the VIIRS's OC and atmospheric data stream through time-series in-situ data acquired at the observatory sites which are part of the AERONET-OC network. This paper addresses the preliminary evaluations of the VIIRS sensor's performance for retrieving OC data of typical coastal water environments, by carrying out time-series, as well as qualitative and quantitative match-up comparisons analysis between in-situ and satellite retrieved OC data. Initial time-series match-up comparisons carried out for a year period (January to December, 2012) show that VIIRS data exhibits strong temporal and statistical agreements with AERONET-OC data demonstrating a potential in enhanced coastal water monitoring from space. VIIRS data of two NASA-OBPG processing schemes which apply different vicarious calibration gains and NOAA-IDPS system are analyzed based on in-situ data of LISCO and WaveCIS AERONET-OC sites which are located in Long Island Sound and Gulf of Mexico respectively as well as OC retrievals of the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor aboard the Aqua satellite. The underlying cause of the discrepancies observed in VIIRS retrieved normalized water-leaving radiances is also investigated. Finally, as the NASA-OBPG and NOAA-IDPS processing schemes for ocean color data of the VIIRS sensor continue to evolve, the results underline the necessity for monitoring and assessing the validity and consistency of VIIRS' ocean color products, especially for coastal waters. [Copyright &y& Elsevier]

The Río de la Plata is one of the largest and most turbid estuaries of the world, carrying a total of 160 million tonsy−1 of suspended sediments. The knowledge of their spatial distribution and their scales of variability is fundamental for management and scientific reasons, but has been limited by the scarcity of observations. During 2009 and 2010, in-situ data (CTD and turbidity profiles, and water and bottom sediment samples) were collected at 26 sites during six repeated cruises and from three fixed instruments deployed in the frame of the FREPLATA/FFEM experiment. In this paper we complement the analysis of this in-situ data base with 10 years of daily intermediate resolution (1km) MODIS-Aqua observations processed for surface suspended matter using the IFREMER algorithm for coastal turbid waters. The aim of this work is to provide a comprehensive characterization of the annual mean suspended matter concentration distribution, to study its variability on seasonal time scale and to identify the involved physical mechanisms. The comparison between the statistics of the direct and remote sensed data is satisfactory, showing a good agreement in the magnitude and spatial distribution of the mean suspended sediments concentration, its standard deviation, so as the seasonal variability. Our data show that all along the year the concentration of surface suspended matter maximizes along the southern coast of the upper and intermediate estuary and at the tips of Samborombón Bay. This fact is linked in part with the higher solid discharge of the Paraná River – flowing along the southern coast – compared to the Uruguay River which flows following the northern coast. The former receives most of the sediments load to the Río de la Plata from the Bermejo River. The observed mean pattern is also related to the stronger tidal currents along the southern coast of the estuary and at the tips of Samborombón Bay, which act re-suspending sediments near the bottom. Then, wind waves during storms enhance vertical mixing, increasing the surface concentration. The concentration of suspended sediments rapidly falls seawards the Barra del Indio shoal, in the area of the salt wedge. In the outer estuary, suspended matter concentration is also strongly associated to the wind-forced motion of the freshwater plume. Suspended matter concentration exhibits a maximum in winter and a minimum in summer, that cannot be fully explained in terms of the seasonal cycle of the solid discharge of the tributaries, but seems to be related to a raise in the frequency of the storms in winter, increasing the frequency of strong winds and higher wind waves, and the associated re-suspension and mixing. [ABSTRACT FROM AUTHOR]

Abstract: Timely and accurate monitoring of forest disturbance is essential to help us understand how the Earth system is changing. MODIS (Moderate Resolution Imaging Spectroradiometer) imagery and subsequent MODIS products provide near-daily global coverage and have transformed the ways we study and monitor the Earth. To monitor forest disturbance, it is necessary to be able to compare observations of the same place from different times, but this is a challenging task using MODIS data as observations from different days have varying view angles and pixel sizes, and cover slightly different areas. In this paper, we propose a method to fuse MODIS and Landsat data in a way that allows for near real-time monitoring of forest disturbance. The method is based on using Landsat time-series images to predict the next MODIS image, which forms a stable basis for comparison with new MODIS acquisitions. The predicted MODIS images represent what the surface should look like assuming no disturbance, and the difference in the spectral signatures between predicted and observed MODIS images becomes the “signal” used for detecting forest disturbance. The method was able to detect subpixel forest disturbance with a producer's accuracy of 81% and a user's accuracy of 90%. Patches of forest disturbance as small as 5 to 7ha in size were detected on a daily basis. The encouraging results indicate that the presented fusion method holds promise for improving monitoring of forest disturbance in near real-time. [Copyright &y& Elsevier]

Abstract: This paper explores the spatiotemporal patterns of total phosphorus (TP) in Tampa Bay (Bay), Florida, with the aid of Moderate Resolution Imaging Spectroradiometer (MODIS) images and genetic programming (GP) models. The study was designed to link TP concentrations with relevant water quality parameters and remote sensing reflectance bands in aquatic environments using in-situ data from a local database to support the calibration and validation of the GP model. The GP models show the effective capacity to demonstrate snapshots of spatiotemporal distributions of TP across the Bay, which helps to delineate the short-term seasonality effects and the decadal trends of TP in an environmentally sensitive coastal bay area. In the past decade, urban development and agricultural activities in the Bay area have substantially increased the use of fertilizers. Landfall hurricanes, including Frances and Jeanne in 2004 and Wilma in 2005, followed by continuous droughts from 2006 to 2008 in South Florida, made the Bay area an ideal place for a remote sensing impact assessment. A changing hydrological cycle, triggered by climate variations, exhibited unique regional patterns of varying TP waste loads into the Bay over different time scales ranging from seasons to years. With the aid of the derived GP model in this study, we were able to explore these multiple spatiotemporal distributions of TP concentrations in the Tampa Bay area aquatic environment and to elucidate these coupled dynamic impacts induced by both natural hazards and anthropogenic perturbations. This advancement enables us to identify the hot moments and hot spots of TP concentrations in the Tampa Bay region. [Copyright &y& Elsevier]

Abstract: In this paper we develop a neural network (NN) algorithm for retrieving inherent optical properties (IOP) from above water remote sensing reflectances (Rrs) at available MODIS (or similar satellite) wavelengths. In previous work we used Hydrolight5 simulations of Rrs with widely varying globally representative constituent physical parameters as a training basis to develop a neural network algorithm, which, using the Rrs at the MODIS visible wavelengths (412, 443, 488, 531, 547 and 667nm) as input, retrieves the in-water particulate backscattering (bbp ), phytoplankton (aph ) and non-phytoplankton (adg ) absorption coefficients at 443nm. In this work, using the same dataset, we develop a NN which takes these same Rrs as input, and produces an output which is used to separate the non-phytoplankton absorption coefficient (adg ) at 443nm into dissolved (ag ) and particulate (adm ) components. We then apply this synthetically trained algorithm to the NASA bio-Optical Marine Algorithm Data set (NOMAD) Rrs to retrieve IOP at 443nm, with the measured NOMAD and retrieved IOP values showing good agreement. These retrieved IOP along with their related Rrs values are then used to train an additional NN that produces chlorophyll concentration [Chl] as output. It is shown that these [Chl] values are retrieved more accurately when compared with ones retrieved with a similar approach which does not use IOP as input, as well as with those derived using the MODIS OC3 algorithm. [Copyright &y& Elsevier]

Abstract: MODIS land surface reflectance product (MOD09) is one of the most popular data sources for characterizing land surface environments. Because those cloudy observations should be excluded from further analysis, the reliable cloud screening is important for downstream applications. In this paper, an approach is proposed to generate cloud masks from time series of MOD09 products. It is found that an inflexion point exists between the clear-sky and cloudy observations if time series of reflectances assembled from the same location are sorted. The maximum surface reflectance can be composited from these inflexions and those observations with reflectance values larger than the inflexion are identified as cloudy. To the best of our knowledge, this is the first method to composite the maximum snow-free surface reflectance. And a new method is proposed to objectively compare cloud detection results derived from different approaches. Comparisons show that this inflexion-based cloud detection algorithm performs generally better than the cloud masks accompanying with the MOD09 products. The new cloud masks are valuable for those applications relying on the MOD09 products as input and for analysis of the uncertainty of the MODIS cloud mask products. [Copyright &y& Elsevier]

Abstract: Land cover maps at different resolutions and mapping extents contribute to modeling and support decision making processes. Because land cover affects and is affected by climate change, it is listed among the 13 terrestrial essential climate variables. This paper describes the generation of a land cover map for Latin America and the Caribbean (LAC) for the year 2008. It was developed in the framework of the project Latin American Network for Monitoring and Studying of Natural Resources (SERENA), which has been developed within the GOFC-GOLD Latin American network of remote sensing and forest fires (RedLaTIF). The SERENA land cover map for LAC integrates: 1) the local expertise of SERENA network members to generate the training and validation data, 2) a methodology for land cover mapping based on decision trees using MODIS time series, and 3) class membership estimates to account for pixel heterogeneity issues. The discrete SERENA land cover product, derived from class memberships, yields an overall accuracy of 84% and includes an additional layer representing the estimated per-pixel confidence. The study demonstrates in detail the use of class memberships to better estimate the area of scarce classes with a scattered spatial distribution. The land cover map is already available as a printed wall map and will be released in digital format in the near future. The SERENA land cover map was produced with a legend and classification strategy similar to that used by the North American Land Change Monitoring System (NALCMS) to generate a land cover map of the North American continent, that will allow to combine both maps to generate consistent data across America facilitating continental monitoring and modeling. [Copyright &y& Elsevier]

Abstract: Africa is home to some of the most vulnerable natural ecosystems and species on the planet. Around 7000 protected areas seek to safeguard the continent''s rich biodiversity, but many of them face increasing management challenges. Human disturbances permeating into the parks directly and indirectly affect the ecological functioning and integrity of protected areas. With the envisaged expansion of the protected area network and further expected population and economic growth in the region, the competition between nature conservation and resources demands is likely to increase. The regular monitoring of land cover in and around protected areas can support the early detection of conservation conflicts. In this paper, we evaluate the use of the annual time series of MODIS Land Cover (LC) type product between 2003 and 2009 to monitor land cover changes at continental scale. We use the mean classification confidence and change frequency as indicators to assess the temporal consistency of the MODIS LC classifier for accurately monitoring land cover changes. We discuss the perspectives and issues for an automated monitoring of land cover changes in African protected areas. [Copyright &y& Elsevier]

Abstract: Atmospheric correction in the Visible and Near Infrared (VNIR) spectral range of remotely sensed data is significantly simplified if we assume a Lambertian target. However, natural surfaces are anisotropic. Therefore, this assumption will introduce an error in surface directional reflectance estimates and consequently in the estimation of vegetation indexes such as the Normalized Difference Vegetation Index (NDVI) and the surface albedo retrieval. In this paper we evaluate the influence of directional effects on the atmospheric correction and its impact in the NDVI and albedo estimation. First, we derived the NDVI and surface albedo from data corrected assuming a Lambertian surface, then by using the BRDF model parameters used in the albedo retrieval, we account for the coupling effect and retrieved an improved NDVI and albedo. The study used Climate Modeling Grid (CMG) MODIS data, which has a spatial resolution of 0.05°. We focused our analysis on four Aeronet sites located in the United States of America: KONZA EDC (a prairie region in the Flint Hills, Kansas), Howland (a forest area in Maine), Walker Branch (a forest area in Tennessee) and GSFC (at NASA''s Goddard Space Flight Center). The results indicate that the relative errors due to the Lambertian assumption on the surface reflectance are 3–12% in the visible and 0.7–5.0% in the near-infrared, around 1% on the NDVI and less than 1% on the albedo. [Copyright &y& Elsevier]

Abstract: This paper presents the assessment of a NDVI-based parametric model (C-Fix) and a bio-geochemical model (BIOME-BGC) for the simulation of semi-natural grassland primary productivity in Italy. The two models are first calibrated using the gross primary productivity (GPP) data of an eddy covariance flux tower placed over a Mediterranean-temperate hilly area in Central Italy. Next, they are applied to estimate the net primary productivity (NPP) of three independent areas representative of different eco-climatic zones. The first area shows a typical Alpine climate, while the other two are characterized by more or less pronounced Mediterranean features. The accuracy of the NPP estimates is assessed through comparison with destructive dry matter measurements taken in the three areas. The results obtained support the capability of the two models to predict spatial NPP differences across the various grassland sites. The greatest estimation errors are found in the mountain area, mostly due to inaccuracies in the meteorological input data. These errors affect particularly the outputs of the bio-geochemical model and are mitigated by the use of C-Fix, which exploits the remotely sensed information related to the seasonal evolution of green biomass. [Copyright &y& Elsevier]

Abstract: Disease forecasting forms an integral part of crop protection for ensuring quality and quantity of production. In this paper, a new method of multi-stage tracking of Sclerotinia rot (Sclerotinia sclerotiorum) disease in a large mustard growing region over 5km×5km (27.00–27.25°N; 77.25–77.50°E) in Bharatpur district of Rajasthan state of North-West India is demonstrated. In addition to surface weather data, post-facto analysis of 5-year (2003–2007) satellite-based data of surface reflectances in red (R), near infrared (NIR) and shortwave infrared (SWIR) bands, land surface temperature (LST) from Moderate Resolution Imaging Spectroradiometer (MODIS) AQUA at day (1:30pm) and night time (1:30am) LST, were done to characterize disease outbreak (stage-I) and persistence (stage-II). While stage-I evaluation was based on anomaly in minimum air temperatures and night time LST, stage-II evaluation was carried out using quadrant-based trapezoidal clusters between soil and canopy dryness indicators. Hyperspectral data on two dates from Hyperion sensor at EO-1 platform were used for two-step spectral discrimination to select bands and disease indices specific to rot. Among all the hyperspectral indices, a three-band rot index (ROTI) was found to be the better one in field scale rot discrimination (stage-III evaluation). The reduction in fractional canopy cover in diseased patches in 2005 as compared to a normal year (2007) indirectly validated the disease effect. [Copyright &y& Elsevier]

Abstract: Remotely sensed data is the main source of vegetation leaf area index (LAI) information on the regional to global scale. Many validation results have revealed that the accuracy of the retrieved LAI is often affected by the cloud cover of imagery, instrument problems, and inversion algorithms. Ground meteorological station data, characterized by relatively high accuracy and time continuity compared with remote sensing data, can provide complementary information to remote sensing observations. In this paper, we combine the potential advantages of both types of data in order to improve LAI retrievals in the Heihe River Basin, an arid and semi-arid area in northwest China where Moderate Resolution Imaging Spectroradiometer (MODIS) LAI values are significantly underestimated. A dynamic Bayesian network (DBN) is used to integrate these two data types for time series LAI estimation. Results show that the square of correlation coefficient between LAI values estimated by our DBN method (referred to as DBN LAI) and field measured LAI values is 0.76, with a root mean square error of 0.78. The DBN LAI are closer to field measurements than the MODIS LAI standard product values. Moreover, by introducing ground meteorological station data using a dynamic process model, DBN LAI show better temporal consistency than the MODIS LAI. It is concluded that the quality of LAI retrievals can be improved by combining remote sensing data and ground meteorological station data using a filtering inference algorithm in a DBN framework. More importantly, the study provides a basis and method for utilizing ground meteorological station network data to estimate land surface parameters on a regional scale. [Copyright &y& Elsevier]

Abstract: Tropical rainforests are significant contributors to the global cycles of energy, water and carbon. As a result, monitoring of the vegetation status over regions such as Amazônia has been a long standing interest of Earth scientists trying to determine the effect of climate change and anthropogenic disturbance on the tropical ecosystems and its feedback on the Earth''s climate. Satellite-based remote sensing is the only practical approach for observing the vegetation dynamics of regions like the Amazon over useful spatial and temporal scales, but recent years have seen much controversy over satellite-derived vegetation states in Amazônia, with studies predicting opposite feedbacks depending on data processing technique and interpretation. Recent results suggest that some of this uncertainty could stem from a lack of quality in atmospheric correction and cloud screening. In this paper, we assess these uncertainties by comparing the current standard surface reflectance products (MYD09, MYD09GA) and derived composites (MYD09A1, MCD43A4 and MYD13A2 — Vegetation Index) from the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard the Aqua satellite to results obtained from the Multi-Angle Implementation of Atmospheric Correction (MAIAC) algorithm. MAIAC uses a new cloud screening technique, and novel aerosol retrieval and atmospheric correction procedures which are based on time-series and spatial analyses. Our results show considerable improvements of MAIAC processed surface reflectance compared to MYD09/MYD13 with noise levels reduced by a factor of up to 10. Uncertainties in the current MODIS surface reflectance product were mainly due to residual cloud and aerosol contamination which affected the Normalized Difference Vegetation Index (NDVI): During the wet season, with cloud cover ranging between 90% and 99%, conventionally processed NDVI was significantly depressed due to undetected clouds. A smaller reduction in NDVI due to increased aerosol levels was observed during the dry season, with an inverse dependence of NDVI on aerosol optical thickness (AOT). NDVI observations processed with MAIAC showed highly reproducible and stable inter-annual patterns with little or no dependence on cloud cover, and no significant dependence on AOT (p <0.05). In addition to a better detection of cloudy pixels, MAIAC obtained about 20–80% more cloud free pixels, depending on season, a considerable amount for land analysis given the very high cloud cover (75–99%) observed at any given time in the area. We conclude that a new generation of atmospheric correction algorithms, such as MAIAC, can help to dramatically improve vegetation estimates over tropical rain forest, ultimately leading to reduced uncertainties in satellite-derived vegetation products globally. [Copyright &y& Elsevier]

Abstract: To deal with water resources crisis caused by climate change and intensified human activities, the Hetao Irrigation District of Inner Mongolia in North China has been experiencing rehabilitation for water-saving since 1999. This has significantly changed its regional hydrological cycle and thus water use patterns over different land use types. In this paper, Surface Energy Balance Algorithm for Land (SEBAL) model fed with MODIS data was applied in Hetao area from 2000 to 2010 to examine the spatial and temporal patterns of evapotranspiration (ET). The SEBAL estimated ET agreed well with that from other methods in the study area. The results indicate that inter-annual variability in ET over agricultural land, water body, woodland and irrigated grassland are primarily explained by the variation of reference ET. So it is with the seasonal variability in agricultural land ET on monthly basis. However, the inter-annual variations of ET over sandy land, Gobi desert and mountain areas are mainly controlled by precipitation. Over the study period, a reduction in river water diversion for irrigation has not reduced the agricultural land ET, indicating no significant impact on agricultural production. But ET over the non-irrigated grassland tends to decrease, which was likely caused by declining groundwater table in recent years. [Copyright &y& Elsevier]