研究目的
Validating a methodology for deriving land-surface energy fluxes using Sentinel-2 and Sentinel-3 observations in a savannah landscape in central Spain.
研究成果
The study demonstrates the feasibility of estimating land surface energy fluxes using Sentinel-2 and Sentinel-3 observations at both coarse and fine scales. The methodology allows for the derivation of fluxes with similar accuracy at both spatial scales, with the fine-scale estimates providing more detailed separation of fluxes originating from individual landscape features. The robustness of the methodology will be further developed and validated in other landscapes and climatic zones.
研究不足
The study is limited to a savannah landscape in central Spain, and the robustness of the methodology needs to be further validated in other landscapes and climatic zones. Additionally, the impact of different meteorological inputs and ET modelling approaches on the accuracy of the modelled fluxes will need further investigation.
1：Experimental Design and Method Selection:
The study uses a Two-Source Energy Balance (TSEB) modelling scheme to estimate land-surface energy fluxes at two spatial resolutions: fine (20m) and coarse (around 1km). Thermal observations from SLSTR on Sentinel-3 and optical observations from MSI on Sentinel-2 are used. For fine resolution estimates, thermal observations are sharpened using high-resolution optical observations and a machine learning algorithm.
2：Sample Selection and Data Sources:
The study site is located at Majadas de Tiétar, Cáceres (Spain) FLUXNET site, a Mediterranean savannah with an overstory dominated by holm oak and understory mainly composed of short herbaceous species.
3：List of Experimental Equipment and Materials:
Sentinel-2 and Sentinel-3 satellite observations, ECMWF ERA5 reanalysis data for meteorological forcing, and a land cover map based on Corine land-cover from
4：Experimental Procedures and Operational Workflow:
20 The methodology involves sharpening coarse-scale LST to fine resolution using a machine learning algorithm, then using the sharpened LST as input to the TSEB model to estimate energy fluxes.
5：Data Analysis Methods:
The accuracy of the modelled fluxes is evaluated against measurements from a flux tower, with accuracy statistics including Root Mean Square Error (RMSE), Bias, coefficient of variation (CV), and correlation (r).
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
