Mapping Microbial Biomass Carbon Dynamics in an Arid Region Under Global Change Using a Novel Remote Sensing-Microbial Model

1. Summary This dataset presents a high-resolution retrieval of Microbial Biomass Carbon (MBC) dynamics in arid and semi-arid regions (case study: Xinjiang, China). The data was generated using a novel Remote Sensing–Microbial Decomposition Process Model, which integrates satellite-derived environmental variables with a Bayesian-optimized microbial process framework. 2. Directory Structure and Content The dataset is organized into two primary folders representing inputs and outputs: I. inputdata observed-data: Field observation data of soil microbial biomass carbon collected from 2022 to 2024. Environmental Variables: Annual gridded datasets used to drive the model, including: GPP (Gross Primary Productivity) LST (Land Surface Temperature) PRE (Precipitation) SM (Soil Moisture) BD (Bulk Density) TMP (Temperature) II. output (Model Results & Post-processing) This directory is divided into two sub-folders: output_model: Contains the year-by-year simulation results of the retrieval model. Each folder contains spatial raster files and attribute data (e.g., .dbf, .shp, .shx for SOC/MBC maps) representing the annual microbial biomass carbon distribution. output_chuli (Post-processed Data): Contains statistical and visual analysis results, including: MBC_yearmean & SOC_yearmean: Multi-year average dynamics. Five year synthetic data: Aggregated datasets for trend analysis. arcgis-drawing: Ready-to-use map layers for visualization. zhifangtu.m: MATLAB script used for generating frequency distribution histograms.