Data and code for: Model-derived nitrogen-response zones identify spatial priorities for N2O mitigation in global wheat and maize croplands

This repository contains the data and Python scripts used in the study of global N2O emission intensity prediction for wheat and maize croplands. Machine learning models (TabPFN, XGBoost, MLP, Ridge Regression) were trained on field-scale N2O observations (1980–2024) and applied to predict global emission intensity under baseline (BAU) and nitrogen reduction scenarios (S2: −20%, S3: −30%). Contents:- data.zip: Field observations, spatial input rasters, and model output rasters- code.zip: All Python analysis scripts- requirements.txt: Python package dependencies- README.txt: Full documentation and usage instructions This study was supported by the International (Regional) Cooperation and Exchange Program (Grant No. 42161144002), the Shandong Provincial Key Research and Development Program (Grant No. 2023TZXD005), and the National Natural Science Foundation of China (Grant No. W2541025).