Remote Sensing Ecological Index (RSEI) raster data for the ecological restoration belt in the southeastern karst region of Southwest China
Description
This dataset comprises annual Remote Sensing Ecological Index (RSEI) raster data for the ecological restoration belt in the southeastern karst region of Southwest China, spanning the period from 2000 to 2022. The dataset has a spatial resolution of 500 meters and is designed to provide a comprehensive assessment of the spatiotemporal dynamics of the Ecological Environment Quality (EEQ) in the study area. The RSEI is a composite index that integrates four key remote sensing indicators through Principal Component Analysis (PCA): Greenness Indicator: Normalized Difference Vegetation Index (NDVI), reflecting vegetation cover. Wetness Indicator: The wetness component (WET) from the Tasseled Cap Transformation, representing the moisture content of soil and vegetation. Dryness Indicator: Normalized Difference Built-up and Soil Index (NDBSI), indicating the degree of surface dryness influenced by built-up areas and bare soil. Heat Indicator: Land Surface Temperature (LST), reflecting the surface thermal environment. All original remote sensing imagery (MOD09A1, MOD13A1, MOD11A2, MOD17A3HGF) was sourced from the National Aeronautics and Space Administration's (NASA) Earth Observing System Data and Information System (EOSDIS). This study utilized the Google Earth Engine (GEE) platform for the systematic processing of these multi-source remote sensing data. The process included image mosaicking, cloud removal, gap-filling, atmospheric correction, and topographic correction to ultimately compute the 23-year annual RSEI time-series data covering the entire study area. This dataset is a core output of the paper titled "Diagnostic Adaptive Assessment Framework (DAAF-Karst): Revealing the Heterogeneous Mechanisms of Ecological Evolution in the Karst Region of Southwest China." It provides a critical data foundation for an in-depth analysis of the ecological restoration and degradation processes, decoupling their complex driving mechanisms, and formulating precise adaptive management strategies for this ecologically fragile region.
Files
Steps to reproduce
The Remote Sensing Ecological Index (RSEI) data for this study were generated using publicly available MODIS imagery on the Google Earth Engine (GEE) platform. Researchers can reproduce the dataset by following these steps: 1. Data Preparation and Preprocessing: Platform: Log in to the Google Earth Engine (GEE) Code Editor. Define Area of Interest (AOI): Upload or define the vector boundary of the study area (e.g., shapefile or GeoJSON). Image Collection Retrieval: Surface Reflectance: Call the MODIS/061/MOD09A1 surface reflectance product for calculating NDVI, WET, and NDBSI. Land Surface Temperature: Call the MODIS/061/MOD11A2 Land Surface Temperature product to obtain LST. Set Time Frame: Define the required years (e.g., 2000 to 2022) and process the data in an annual loop. Image Preprocessing: Write a GEE script to perform cloud masking, gap-filling, mosaicking, and clipping on the annual image collections to generate a cloud-free composite for the study area for each year. 2. RSEI Indicator Calculation: Greenness (NDVI): Calculate NDVI using the Near-Infrared (NIR) and Red bands. Formula: NDVI = (NIR - Red) / (NIR + Red). Wetness (WET): Compute the wetness component using the Tasseled Cap Transformation coefficients applicable to MODIS sensor data, based on the surface reflectance bands. Dryness (NDBSI): Calculated as the average of the Soil Index (SI) and the Index-Based Built-up Index (IBI), which are also derived from the corresponding MODIS bands. Heat (LST): Extract LST data from the MOD11A2 product, performing necessary unit conversions and scaling to standardize the units. 3. Indicator Normalization: To address the different units and ranges of the four indicators (NDVI, WET, NDBSI, LST), normalize each to a scale of 0 to 1. Note that WET and NDVI are positive indicators, while NDBSI and LST are negative indicators. The direction of the negative indicators must be reversed (e.g., by subtracting the normalized value from 1) before proceeding. 4. Principal Component Analysis (PCA) and RSEI Generation: Combine the four normalized indicators into a single 4-band image. Apply Principal Component Analysis (PCA) to this composite image. The first principal component (PC1), which typically captures the majority of the variance from all indicators, is extracted. Use the extracted PC1 as the initial RSEI value (RSEI_initial). 5. Final RSEI Normalization and Export: Perform a final normalization on the initial RSEI values (RSEI_initial) to scale them to a final range of 0 to 1. Formula: RSEI = (RSEI_initial - RSEI_min) / (RSEI_max - RSEI_min). Export the final annual RSEI raster data from GEE, typically in GeoTIFF format. For detailed calculation formulas and parameter settings, please refer to the supplementary materials of the main paper. Following these steps will allow for the replication of the annual RSEI dataset used in this research.
Institutions
- Yunnan Normal University