Articles | Volume 14, issue 3
Model evaluation paper
16 Mar 2021
Model evaluation paper |  | 16 Mar 2021

Using Shapley additive explanations to interpret extreme gradient boosting predictions of grassland degradation in Xilingol, China

Batunacun, Ralf Wieland, Tobia Lakes, and Claas Nendel

Related authors

Evaluation and optimisation of the soil carbon turnover routine in the MONICA model (version 3.3.1)
Konstantin Aiteew, Jarno Rouhiainen, Claas Nendel, and Rene Dechow
EGUsphere,,, 2023
Short summary
Aerial and surface rivers: downwind impacts on water availability from land use changes in Amazonia
Wei Weng, Matthias K. B. Luedeke, Delphine C. Zemp, Tobia Lakes, and Juergen P. Kropp
Hydrol. Earth Syst. Sci., 22, 911–927,,, 2018
Short summary
Climate impacts on human livelihoods: where uncertainty matters in projections of water availability
T. K. Lissner, D. E. Reusser, J. Schewe, T. Lakes, and J. P. Kropp
Earth Syst. Dynam., 5, 355–373,,, 2014
Short summary

Related subject area

Earth and space science informatics
Machine learning for numerical weather and climate modelling: a review
Catherine O. de Burgh-Day and Tennessee Leeuwenburg
Geosci. Model Dev., 16, 6433–6477,,, 2023
Short summary
Overcoming barriers to enable convergence research by integrating ecological and climate sciences: the NCAR–NEON system Version 1
Danica L. Lombardozzi, William R. Wieder, Negin Sobhani, Gordon B. Bonan, David Durden, Dawn Lenz, Michael SanClements, Samantha Weintraub-Leff, Edward Ayres, Christopher R. Florian, Kyla Dahlin, Sanjiv Kumar, Abigail L. S. Swann, Claire M. Zarakas, Charles Vardeman, and Valerio Pascucci
Geosci. Model Dev., 16, 5979–6000,,, 2023
Short summary
Ensemble of optimised machine learning algorithms for predicting surface soil moisture content at a global scale
Qianqian Han, Yijian Zeng, Lijie Zhang, Calimanut-Ionut Cira, Egor Prikaziuk, Ting Duan, Chao Wang, Brigitta Szabó, Salvatore Manfreda, Ruodan Zhuang, and Bob Su
Geosci. Model Dev., 16, 5825–5845,,, 2023
Short summary
Hazard assessment modeling and software development of earthquake-triggered landslides in the Sichuan–Yunnan area, China
Xiaoyi Shao, Siyuan Ma, and Chong Xu
Geosci. Model Dev., 16, 5113–5129,,, 2023
Short summary
A generalized spatial autoregressive neural network method for three-dimensional spatial interpolation
Junda Zhan, Sensen Wu, Jin Qi, Jindi Zeng, Mengjiao Qin, Yuanyuan Wang, and Zhenhong Du
Geosci. Model Dev., 16, 2777–2794,,, 2023
Short summary

Cited articles

Abdullah, A. Y. M., Masrur, A., Adnan, M. S. G., Baky, Md. A. A., Hassan, Q. K., and Dewan, A.: Spatio-temporal Patterns of Land Use/Land Cover Change in the Heterogeneous Coastal Region of Bangladesh between 1990 and 2017, Remote Sens., 11, 790,, 2019. 
Aburas, M. M., Ahamad, M. S. S., and Omar, N. Q.: Spatio-temporal simulation and prediction of land-use change using conventional and machine learning models: a review, Environ. Monit. Assess., 191,, 2019. 
Abu-Rmileh, A.: Be careful when interpreting your features importance in XGBoost!, Data Sci., available at:, last access: 14 June 2019. 
Ahmadlou, M., Delavar, M. R., and Tayyebi, A.: Comparing ANN and CART to Model Multiple Land Use Changes: A Case Study of Sari and Ghaem-Shahr Cities in Iran, J. Geomat. Sci. Technol., 6, 292–303, 2016. 
Ahmadlou, M., Delavar, M. R., Basiri, A., and Karimi, M.: A Comparative Study of Machine Learning Techniques to Simulate Land Use Changes, J. Indian Soc. Remote Sens., 47, 53–62,, 2019. 
Short summary
Extreme gradient boosting (XGBoost) can provide alternative insights that conventional land-use models are unable to generate. Shapley additive explanations (SHAP) can interpret the results of the purely data-driven approach. XGBoost achieved similar and robust simulation results. SHAP values were useful for analysing the complex relationship between the different drivers of grassland degradation.