Moderate Resolution Imaging Spectroradiometer Albedo Products Research Articles

Estimating 10-m land surface albedo from Sentinel-2 satellite observations using a direct estimation approach with Google Earth Engine

Land surface albedo plays an important role in controlling the surface energy budget and regulating the biophysical processes of natural dynamics and anthropogenic activities. Satellite remote sensing is the only practical approach to estimate surface albedo at regional and global scales. It nevertheless remains challenging for current satellites to capture fine-scale albedo variations due to their coarse spatial resolutions from tens to hundreds of meters. The emerging Sentinel-2 satellites, with a high spatial resolution of 10 m and an approximate 5-day revisiting cycle, provide a promising solution to address these observational limitations, yet their potentials remain underexplored. In this study, we integrated the Sentinel-2 observations with an updated direct estimation approach to improve the estimation and monitoring of fine-scale surface albedo. To enable the capability of the direct estimation approach at a 10-m scale, we combined the 10-m resolution European Space Agency (ESA) WorldCover land cover data and the 500-m resolution Moderate-Resolution Imaging Spectroradiometer (MODIS) Bidirectional Reflectance Distribution Function (BRDF)/albedo product to build a high-quality and representative BRDF training database. To evaluate our approach, we proposed an integrated evaluation framework leveraging 3-D physical model simulations, ground measurements, and satellite observations. Specifically, we first simulated a comprehensive dataset of Sentinel-2-like surface reflectance and broadband albedo across a variety of geometric configurations using the MODIS BRDF training samples. With this dataset, we built the Look-Up-Tables (LUTs) that connect surface broadband albedo and Sentinel-2 reflectance through a direct angular bin-based linear regression approach, and further coupled these LUTs with the Google Earth Engine (GEE) cloud-computing platform. We next evaluated the proposed algorithm at two spatial levels: (1) 10-m scale for absolute accuracy assessment using the references from the Discrete Anisotropic Radiative Transfer (DART) simulations and flux-site observations, and (2) 500-m scale for large-scale mapping assessment by comparing the estimated albedo with the MODIS albedo product. Lastly, we presented four examples to show the capability of Sentinel-2 albedo in detecting fine-scale characteristics of vegetation and urban covers. Results show that: (1) the proposed algorithm accurately estimates surface albedo from Sentinel-2-like reflectance across different landscape configurations (overall root-mean-square-error (RMSE) = 0.018, bias = 0.005, and coefficient of determination (R2) = 0.88); (2) the Sentinel-2-derived surface albedo agrees well with ground measurements (overall RMSE = 0.030, bias = -0.004, and R2 = 0.94) and MODIS products (overall RMSE = 0.030, bias = 0.021, and R2 = 0.97); and (3) Sentinel-2-derived albedo accurately captures seasonal leaf phenology and rapid snow events, and detects the interspecific (or interclass) variations of tree species and colored urban rooftops. These results demonstrate the capability of the proposed approach to map high-resolution surface albedo from Sentinel-2 satellites over large spatial and temporal contexts, suggesting the potential of using such fine-scale datasets to improve our understanding of albedo-related biophysical processes in the coupled human-environment system.

Simulated cold bias being improved by using MODIS time-varying albedo in the Tibetan Plateau in WRF model

Systematic cold biases exist in the simulation for 2 m air temperature in the Tibetan Plateau (TP) when using regional climate models and global atmospheric general circulation models. We updated the albedo in the Weather Research and Forecasting (WRF) Model lower boundary condition using the Global LAnd Surface Satellite Moderate-Resolution Imaging Spectroradiometer albedo products and demonstrated evident improvement for cold temperature biases in the TP. It is the large overestimation of albedo in winter and spring in the WRF model that resulted in the large cold temperature biases. The overestimated albedo was caused by the simulated precipitation biases and over-parameterization of snow albedo. Furthermore, light-absorbing aerosols can result in a large reduction of albedo in snow and ice cover. The results suggest the necessity of developing snow albedo parameterization using observations in the TP, where snow cover and melting are very different from other low-elevation regions, and the influence of aerosols should be considered as well. In addition to defining snow albedo, our results show an urgent call for improving precipitation simulation in the TP.

Spatial scaling of reflectance and surface albedo over a mixed-use, temperate forest landscape during snow-covered periods

Albedo products from the MODerate resolution Imaging Spectroradiometer (MODIS) have been validated extensively over spatially homogeneous sites with snow-cover. This study evaluated the spatial scaling of albedo and related reflectance-based quantities over areas of high spatial heterogeneity in temperate mixed forest, deciduous forest, urban and built-up, and cropland/natural mosaic lands under snow covered conditions. Reflectance-based quantities evaluated included spectral radiance, surface directional reflectance, and spectral and broadband albedo derived from ground- and tower-based measurements and high-resolution (5m) hyperspectral imagery (HSI) to coarse resolution (~500m) MODIS satellite data.Our approach first compared ground- and tower-based spectral and broadband reflectance quantities to HSI data, then to evaluate HSI with MODIS reflectance-based products. Over snow-covered pasture, HSI directional reflectance was biased lower than ground-based measurements collected using a spectroradiometer, and greatly underestimated at wavelengths less than 450nm. Tower-based shortwave broadband albedo (280–2800nm) and HSI albedo agreed within ±0.04. HSI directional spectral reflectance agreed well with tower-based measurements of spectral albedo collected using a spectroradiometer and remote cosine receptor above a mixed forest canopy with underlying snow cover. Spectral albedo collected over a dormant deciduous broadleaf canopy increased from 0.10 for snow-free conditions to 0.14–0.18 when snow-cover was present under the canopy. Canopy shortwave broadband albedo was not very sensitive to underlying snowpack depth, indicative of strong vegetation masking.Next, HSI data were spatially aggregated and averaged to 500m MODIS grids and compared to two MODIS albedo products: (1) MODIS/Terra and Aqua Combined Bidirectional Reflectance Distribution Function (BRDF) Albedo V005 (MCD43A, magnitude inversion), and (2) MODIS/Terra Snow Cover Daily L3 Global 500m Grid (MOD10A1). An assessment of surface heterogeneity demonstrated that MODIS products generally perform well for snow-covered landscapes with high spatial heterogeneity, with biases between ±0.04 and RMSEs less than 0.085 for mixed and deciduous broadleaf forested and urban & built-up land cover classes, and less than 0.11 for cropland/natural mosaic. Biases were generally lower for MOD10A1 compared to MCD43A3.Mean shortwave broadband albedo from MODIS and HSI over deciduous broadleaf and mixed forest with underlying snow cover ranged from 0.14 to 0.28. Albedo over MODIS grids characterized as a mosaic of cropland, forests, grasslands, and shrublands (e.g., cropland/natural mosaic) was substantially higher than forested grids under snow-covered conditions, ranging from 0.39 to 0.43.

Direct estimation of land surface albedo from VIIRS data: Algorithm improvement and preliminary validation

AbstractLand surface albedo (LSA), part of the Visible Infrared Imaging Radiometer Suite (VIIRS) surface albedo environmental data record (EDR), is an essential variable regulating shortwave energy exchange between the land surface and the atmosphere. Two sub‐algorithms, the dark pixel sub‐algorithm (DPSA) and the bright pixel sub‐algorithm (BPSA), were proposed for retrieving LSA from VIIRS data. The BPSA estimates LSA directly from VIIRS top‐of‐atmosphere (TOA) reflectance through simulation of atmospheric radiative transfer. Several changes have been made to improve the BPSA since the deployment of VIIRS. A database of the Moderate Resolution Imaging Spectroradiometer (MODIS) bidirectional reflectance distribution function (BRDF) is collected and converted to bidirectional reflectance at VIIRS bands. The converted reflectance is then used as input to the atmospheric radiative transfer model to generate a look‐up table (LUT) of regression coefficients with consideration of surface BRDF. Before its implementation in the operational system, the new BPSA is tested on the local infrastructure. The incorporation of the surface BRDF improves the accuracy of LSA estimation and reduces the temporal variation of LSA over stable surfaces. VIIRS LSA retrievals agree well with the MODIS albedo products. Comparison with field measurements at seven Surface Radiation (SURFRAD) Network sites shows that VIIRS LSA retrieved from the LUT with surface BRDF has an R2 value of 0.80 and root mean square error of 0.049, better than MODIS albedo products. The VIIRS results have a slight negative bias of 0.004, whereas the MODIS albedo is underestimated with a larger negative bias of 0.026.

Retrieval of aerosol optical depth and surface reflectance over land from NOAA AVHRR data

An algorithm for the land aerosol and bidirectional reflectance inversion by times series technique (LABITS) is proposed and applied to data from the National Oceanic and Atmospheric Administration (NOAA) advanced very high resolution radiometer (AVHRR). The surface reflectance and aerosol optical depth are inverted from AVHRR Channel 1 data using a model for the earth-atmosphere system which couples an atmospheric radiative transfer model with the Ross–Thick–Li-sparse bidirectional reflectance factor (BRF) model. Basic assumptions are that the surface bidirectional reflective property does not vary over a 2–4day periods and that the aerosol characteristics are uniform within a 0.1°×0.1° window (approximately 10km×10km). The LABITS algorithm is applied to data from AVHRR on the NOAA-15, NOAA-16, and NOAA-18 satellites over four distinct areas, namely, North America, Europe, the Sahara and India, to simultaneously retrieve the aerosol optical depth (AOD), BRF parameters and surface albedo. Preliminary results show that AOD and reflectance retrieved from the three different instruments are in good agreement and that LABITS provides good results over both bright surfaces, e.g. the Sahara, and dark surfaces, e.g. Europe. Evaluation of the AOD versus data from the Aerosol Robotic Network (AERONET) provides a correlation coefficient R2 of 0.88 and a root-mean-square error (RMSE) of approximately 0.07; and the uncertainty is approximately Δτ=±0.05±0.20τ. Comparing our results with the moderate resolution imaging spectroradiometer (MODIS) AOD products, over many areas, provides biases in the range of ±0.05. The surface albedo values calculated from the retrieved BRF parameters are similar to those provided by the MODIS albedo product (MCD43). The robustness and applicability of the LABITS algorithm are demonstrated with the retrieval of AOD over China during August 2008. Daily and monthly averaged results show good agreement with collocated AERONET observations and AQUA MODIS products (MYD04 and MYD08). The AOD uncertainty is estimated as Δτ=±0.05±0.30τ. The preliminary analysis of time series over selected AERONET sites shows that the temporal variations of the AOD values retrieved by application of LABITS to AVHRR data are overall similar to temporal variations of AOD provided by the MODIS and AERONET. The algorithm has the potential to retrieve global AOD over land for long time series of NOAA AVHRR data going back to the 1980s, which are urgently needed for studies on aerosol climatology and global climate change.

Evaluation of surface albedo from GEWEX-SRB and ISCCP-FD data against validated MODIS product over the Tibetan Plateau

[1] This paper presents a scheme to examine the accuracy of albedos used in two operational satellite-based radiation data sets over the Tibetan Plateau (TP): the Global Energy and Water Cycle Experiment-Surface Radiation Budget (GEWEX-SRB) and the International Satellite Cloud Climatology Project-Flux Data (ISCCP-FD). First, albedo observations at six sites are used to validate the Moderate Resolution Imaging Spectroradiometer (MODIS) 16 day albedo product with a 500 m spatial resolution. It is shown that both the MODIS black-sky and white-sky albedo products are comparable to the in situ measurements made during snow-free periods. Then, the MODIS 500 m albedos are taken as the ground truth, aggregated to the coarse-resolution grid cells, and compared with the GEWEX-SRB and ISCCP-FD albedo values. The results indicate that both data sets have negative biases in the central and western TP and large positive biases in the southeastern glaciated region. It is recommended that the MODIS albedo product be applied as the input in producing the GEWEX-SRB and ISCCP-FD data sets, at least when processing summer data over the TP. However, the MODIS albedo algorithm needs to be improved to take the effect of snow into consideration when the snow duration is much shorter than the MODIS 16 day inversion window.

Variability of Albedo and Utility of the MODIS Albedo Product in Forested Wetlands

Albedo was monitored over a two-year period (beginning April 2008) at three forested wetland sites in Florida, USA using up- and down-ward facing pyranometers. Water level, above and below land surface, is the primary control on the temporal variability of daily albedo. Relatively low reflectivity of water accounts for the observed reductions in albedo with increased inundation of the forest floor. Enhanced canopy shading of the forest floor was responsible for lower sensitivity of albedo to water level at the most dense forest site. At one site, the most dramatic reduction in daily albedo was observed during the inundation of a highly-reflective, calcareous periphyton-covered land surface. Satellite-based Moderate-Resolution Imaging Spectroradiometer (MODIS) estimates of albedo compare favorably with measured albedo. Use of MODIS albedo values in net radiation computations introduced a root mean squared error of less than 4.7 W/m2 and a mean, annual bias of less than 2.3 W/m2 (1.7%). These results suggest that MODIS-estimated albedo values can reliably be used to capture areal and temporal variations in albedo that are important to the surface energy balance.

Validation of MODIS Albedo Products of Paddy Fields in Japan

A study was conducted in Chiba, Japan, to validate Moderate Resolution Imaging Spectroradiometer (MODIS) albedo products by taking the field measurements of shortwave band albedos in paddy fields. A large difference in spatial scale, from field-measured point data to 1-km resolution, complicates the validation process. To assess such effect of different spatial scales, Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and Enhanced Thematic Mapper Plus (ETM+) data were used. Spatial scale effects on the albedo were examined from three viewpoints: 1) comparison between point-based albedo and mean of albedo in homogeneous area; 2) comparison between point-based albedo and 1-km aggregated albedo; and 3) assessment of semivariogram of albedo in homogeneous area. In implementation of viewpoint 2), Liang's regression model was applied to convert ASTER reflectance into shortwave band albedo. The 1-km ASTER albedo was estimated using the point spread function, and in the same manner, 1-km ETM+ albedo was estimated. All results represent that an area around the measurement site can be assumed to be homogeneous, indicating negligible effects of spatial resolution difference during most of the periods. Comparison of ground-point-based albedos with MODIS actual albedo, estimated from MODIS black-sky albedo, white-sky albedo, and a fraction of diffuse skylight, showed that the accuracy of MODIS albedo products for paddy fields in Japan is within approximately 0.026 by absolute value (root-mean-square error) and 15.1% by relative value

Characterizing the Spatial Variability of Broadband Albedo in a Semidesert Environment for MODIS Validation

Global data sets on land surface albedo will be one of the core products to be derived from data acquired by the Moderate Resolution Imaging Spectroradiometer (MODIS), part of NASA's Earth Observing System. Widespread acceptance of this product by the user communities is dependent, in part, on a comprehensive and rigorous programme of calibration and validation. Since the MODIS albedo product will be produced at a spatial resolution of 1 km, while measurements obtained from field instruments typically relate to areas of only a few tens of meters, this requires an understanding of the spatial variability of land surface albedo and a robust means of scaling up from field to satellite measurements. In this article, we examine these issues for a semidesert environment (the PROVE'97 field site at Jornada, New Mexico, USA). Spatial variations in field measurements of broadband albedo are related to the fractional ground cover of different scene elements (live and senescent vegetation, soil and shadow) via a simple linear mixture model. Information on the fractional ground cover of the scene elements is derived from ground-based hemispherical photography. It is shown that the albedo values predicted by the mixture model are accurate to within 2% of the corresponding measured values. This approach offers considerable potential for the validation of MODIS-derived albedo values through the use of spectral mixture modelling applied to fine spatial resolution satellite sensor images.