LULC_GAN and RSEI_GAN training datasets and multi-scenario prediction results (2000–2035)

This dataset accompanies the paper “Spatiotemporal Prediction and Analysis of LULC Change and Ecological Impact based on Coupled Generative Adversarial Networks.” It contains two main components:(1) Model_training_datasets.tar – includes image tiles and associated metadata used for training and validating the LULC_GAN and RSEI_GAN models;(2) Prediction_results.rar – includes model-generated Land Use/Land Cover (LULC) and Remote Sensing Ecological Index (RSEI) prediction maps for Shanghai from 2020 to 2035, as well as simulation results under five alternative development scenarios (Baseline, Ecological Protection, Agricultural Protection, Urban Development, and Balanced Development). All files are provided in open-access format under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.