Spatiotemporal dynamics of land surface temperature in Yunnan, China: Interpreting multi-temporal drivers using eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanations (SHAP)

doi:10.1016/j.regsus.2026.100350

Abstract

Abstract:

Amid global warming, mountainous regions have emerged as critical zones of investigation owing to their heightened vulnerability to climate change, their ecological significance, and the intensified interactions between natural stress and human activities. Land surface temperature (LST) is a fundamental indicator for assessing climatic sensitivity in these landscapes. However, a comprehensive understanding of the spatiotemporal dynamics and driving mechanisms of LST across large mountainous regions remains limited. Therefore, data from the Terra Moderate Resolution Imaging Spectroradiometer Land Surface Temperature/Emissivity Daily (MOD11A1) Version 6.1 product during 2001-2020 in Yunnan Province (a complex mountainous region), China, were analyzed. Sen’s slope analysis and Mann-Kendall test were applied to detect LST trends and spatial heterogeneity at both annual and seasonal scales. Subsequently, an eXtreme Gradient Boosting (XGBoost) model coupled with SHapley Additive exPlanations (SHAP) was employed to clarify the nonlinear contributions of multiple drivers. The study revealed the following findings. LST exhibited an overall warming rate of 0.020°C/a, characterized by daytime cooling (-0.008°C/a) and nighttime warming (0.048°C/a). LST increased during spring, summer, and autumn (0.011°C/a-0.018°C/a), whereas winter LST exhibited a cooling trend (-0.011°C/a). These variations were spatially partitioned by the Ailao Mountains, with the southwest displaying stronger thermal changes than the northeast. Natural controls, including digital elevation model (DEM) and downward shortwave radiation (DSR), predominated in the northwest high mountain canyons area and south tropical rainforest area, whereas nature-human interactions were more pronounced in the central urban agglomeration warming area and southeast karst landform area. The dominant drivers consisted of DEM, DSR, Normalized Difference Moisture Index, particulate matter 2.5 (PM_2.5), and aerosol optical depth (AOD). The strong correlations between gross domestic product and population density (correlation coefficient (r)=0.95), as well as between PM_2.5 and AOD (r=0.84), highlighted the increasing influence of socioeconomic factors on surface warming. This study can advance the understanding of how mountain topography, moisture, and anthropogenic pressures jointly regulate surface thermal regimes and provide region-specific insights for climate adaptation and sustainable ecosystem management.

Key words: Land surface temperature (LST), Sen’s slope analysis, Mann-Kendall test, Driving mechanism eXtreme Gradient Boosting (XGBoost), SHapley Additive exPlanations (SHAP) Yunnan

LIAO Chaolian, HE Xudong, LONG Xiaomin, WAN Ailing, ZHOU Ruliang, WANG Yanxia. Spatiotemporal dynamics of land surface temperature in Yunnan, China: Interpreting multi-temporal drivers using eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanations (SHAP)[J]. Regional Sustainability, 2026, 7(3): 100350.

Figures/Tables 11

Fig. 1.

Table 1

Table 2

Table 3

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

Fig. 6.

Fig. 7.

Fig. 8.

References 58

[1]	China Meteorological Administration, 2025. China Climate Change Blue Book (2025). [2025-04-10]. https://www.cma.gov.cn/2011xwzx/2011xqxxw/2011xqxyw/202506/t20250630_7175137.html (in Chinese).
[2]	Ding, W.X., Wu, Q., Chen, Y.X., 2024. Trends of maximum annual sea surface temperature in the eastern China seas. Frontiers in Marine Science. 11, 1452125, doi: 10.3389/fmars.2024.1452125.
[3]	Du, J.Z., Wang, K.C., Cui, B.S., et al., 2020. Correction of inhomogeneities in observed land surface temperatures over China. Journal of Climate. 33(20), 8885-8902.
[4]	Fan, Z.Y., Zhang, Q.M., Liu, H.M., et al., 2019. Spatial-temporal distribution of urban heat island and the heating effect of impervious surface in summer in Wuhan. Journal of Geo-information Science. 21(2), 226-235 (in Chinese).
[5]	Grekousis, G., 2025. Geographical-XGBoost: a new ensemble model for spatially local regression based on gradient-boosted trees. Journal of Geographical Systems. 27, 169-195.
[6]	Hamed, K.H., 2008. Trend detection in hydrologic data: The Mann-Kendall trend test under the scaling hypothesis. Journal of Hydrology. 349(3-4), 350-363.
[7]	He, Z.W., Tang, B.H., 2023. Spatiotemporal change patterns and driving factors of land surface temperature in the Yunnan-Kweichow Plateau from 2000 to 2020. Science of The Total Environment. 896, 165288, doi: 10.1016/j.scitotenv.2023.165288.
[8]	Hu, D., Guo, F., Meng, Q., et al., 2024. A novel dual-layer composite framework for downscaling urban land surface temperature coupled with spatial autocorrelation and spatial heterogeneity. International Journal of Applied Earth Observation and Geoinformation. 130, 103900, doi: 10.1016/j.jag.2024.103900.
[9]	Huang, B.Z., He, B.J., 2025. Lawn and irrigation cooling from ground longwave radiation reduction: Understanding the climate-driven variability in cooling performance. Urban Climate. 60, 102360, doi: 10.1016/j.uclim.2024.102360.
[10]	IPCC (Intergovernmental Panel on Climate Change), 2023. Climate Change 2023: Synthesis Report. In:Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Geneva: IPCC.
[11]	Jia, R.Y., Liu, J.W., He, T., et al., 2024. Population heat exposure risk from the perspective of urban heat island spatial expansion in China during 2005-2020. Urban Climate. 56, 101987, doi: 10.1016/j.uclim.2024.101987.
[12]	Kang, L.G., Cao, S.Q., Cao, G.C., et al., 2023. Spatiotemporal variation characteristics of LST in the Qinghai Lake Basin. Arid Land Geography. 46(7), 1084-1097 (in Chinese).
[13]	Li, B., Wang, H.M., Qin, M.Z., et al., 2017a. Comparative study on the relationship between NDVI, NDMI, and LST. Progress in Geography. 36(5), 585-596 (in Chinese).
[14]	Li, Q.X., Sun, W.B., Huang, B.Y., et al., 2020. Consistency of global warming trends strengthened since 1880s. Science Bulletin. 65(20), 1709-1712.
[15]	Li, Q.X., Sun, W.B., Yun, X., et al., 2021. An updated evaluation of the global mean land surface air temperature and surface temperature trends based on CLSAT and CMST. Climate Dynamics. 56(1-2), 635-650.
[16]	Li, X.M., Zhou, Y.Y., Asrar, G.R., et al., 2017b. The surface urban heat island response to urban expansion: a panel analysis for the conterminous United States. Science of The Total Environment. 605-606, 426-435.
[17]	Li, Y.F., Yang, K., Yang, R., 2016. Land surface temperature estimation from remote sensing data—A case study in Kun Ming City. In:Proceedings of the 2015 23^rd International Conference on Geoinformatics. Wuhan, China.
[18]	Liang, H.W., Alimujiang, K.S.M., Zhang, X.L., et al., 2023. Spatiotemporal changes in LST and their influencing factors in oasis urban agglomerations in arid regions: A case study of the urban agglomeration on the northern slope of the Tianshan Mountains. Acta Ecologica Sinica. 43(9), 3650-3664 (in Chinese).
[19]	Liao, C.L., Zhou, P.F., Ye, J.X., et al., 2025. Spatial and temporal distribution of land surface temperature and its driving factors in Central Yunnan. Journal of Guangxi Normal University (Natural Science Edition). 43(1), 185-200 (in Chinese).
[20]	Lin, Z.L., Xu, H.Q., Han, L.Y., et al., 2024. Day and night: impact of 2D/3D urban features on land surface temperature and their spatiotemporal non-stationary relationships in urban building spaces. Sustainable Cities and Society. 108, 103829, doi: 10.1016/j.scs.2024.103829.
[21]	Liu, B.Q., Wu, G.X., Mao, JY., et al., 2013. Genesis of the South Asian High and its impact on the Asian summer monsoon onset. Journal of Climate. 26(9), 2976-2991.
[22]	Liu, P., Wu, G.J., Cao, B., et al., 2024. Variation in glacier albedo on the Tibetan Plateau between 2001 and 2022 based on MODIS data. Remote Sensing. 16(18), 3472, doi: 10.3390/rs16183472.
[23]	Liu, X., He, J.T., Xiong, K., et al., 2022. Identification of factors affecting public willingness to pay for heat mitigation and adaptation: Evidence from Guangzhou, China. Urban Climate. 48, 101405, doi: 10.1016/j.uclim.2022.101405.
[24]	Lundberg, S., Lee, S., 2017. A unified approach to interpreting model predictions. In:Proceedings of the 31^st International Conference on Neural Information Processing Systems (NIPS 2017). Red Hook: Curran Associates Inc., 4768-4777.
[25]	Luo, X., Yang, J., Sun, W., et al., 2021. Suitability of human settlements in mountainous areas from the perspective of ventilation: A case study of the main urban area of Chongqing. Journal of Cleaner Production. 310, 127467, doi: 10.1016/j.jclepro.2021.127467.
[26]	Ma, X.F., Huang, G., Cao, J.J., 2022. The significant roles of anthropogenic aerosols on surface temperature under carbon neutrality. Science Bulletin. 67(5), 470-473.
[27]	Mao, K.B., Ma, Y., Tan, X.L., et al., 2016. Global surface temperature change analysis based on MODIS data in recent twelve years. Advances in Space Research. 59(4), 503-512.
[28]	Mao, K.B., Yan, Y.B., Zhao, B., et al., 2023. Spatiotemporal variation and driving factors of land surface temperature in China. Journal of Catastrophology. 38(2), 60-73 (in Chinese).
[29]	Peng, S.Z.,2020. 1-km Monthly Precipitation Dataset for China (1901-2024). [2025-04-10]. https://www.tpdc.ac.cn/en/data/faae7605-a0f2-4d18-b28f-5cee413766a2 (in Chinese).
[30]	Ran, Y.J., Lei, D.M., Liu, L., et al., 2021. Impact of land use change on ecosystem service value in urban agglomeration of central Yunnan Province during 2000-2020. Bulletin of Soil and Water Conservation. 41(4), 310-322 (in Chinese).
[31]	Ren, J.Y., Yang, J., Yu, W.B., et al., 2024. Investigating the attribution of urban thermal environment changes under background climate and anthropogenic exploitation scenarios. Sustainable Cities and Society. 107, 105466, doi: 10.1016/j.scs.2024.105466.
[32]	Skiles, M.K., Painter, T.H., Deems, J.S., et al., 2013. Dust radiative forcing in snow of the upper Colorado River basin: 2. Interannual variability in radiative forcing and snowmelt rates. Water Resources Research. 48(7), WR011986, doi: 10.1029/2012WR011986.
[33]	Song, Z., Li, R., Qiu, R., et al., 2018. Global land surface temperature influenced by vegetation cover and PM_2.5 from 2001 to 2016. Remote Sensing. 10(12), 2034, doi: 10.3390/rs10122034.
[34]	Taripanah, F., Ranjbar, A., 2021. Quantitative analysis of spatial distribution of land surface temperature (LST) in relation to ecohydrological, terrain and socio-economic factors based on Landsat data in mountainous area. Advances in Space Research. 68(9), 3622-3640.
[35]	Tian, H., Liu, L., Zhang, Z.Y., et al., 2024. Spatiotemporal differentiation and attribution of land surface temperature in China in 2001-2020. Journal of Geographical Sciences. 34(2), 375-396 (in Chinese).
[36]	Tian, Q., Prange, M., Merkel, U., 2016. Precipitation and temperature changes in the major Chinese river basins during 1957-2013 and links to sea surface temperature. Journal of Hydrology. 536, 208-221.
[37]	Tun, Z.K., Pramanik, M., Mallick, S.K., et al., 2025. Cooling the cities: A comprehensive review of urban heat island mitigation strategies in Southeast Asia. Human Settlements and Sustainability. 1(2), 91-102.
[38]	Wang, A., Dai, Y., Zhang, M.M., et al., 2025a. Exploring the cooling intensity of green cover on urban heat island: A case study of nine main urban districts in Chongqing. Sustainable Cities and Society. 124, 106299, doi: 10.1016/j.scs.2024.106299.
[39]	Wang, J.C., Dong, L.J., Ji, S.Y., 2025b. Rock mass instability early warning model: A case study of a high and steep annular slope mining areas using Sen’s slope trend analysis. Tunnelling and Underground Space Technology. 159, 106514, doi: 10.1016/j.tust.2025.106514.
[40]	Wang, X.Y., Yang, J., Yu, W.B., et al., 2024. Spatial effect of urban morphology on land surface temperature from the perspective of local climate zone. Remote Sensing Applications: Society and Environment. 36, 101324, doi: 10.1016/j.rsase.2024.101324.
[41]	Wang, Y.X., Ding, K., Huang, X.Y., et al., 2014. Study on the temperature lapse rate in Yunnan mountainous areas using remote sensing instantaneous temperature fields. Journal of Remote Sensing. 18(4), 912-922 (in Chinese).
[42]	Wei, J., Li, Z.Q., 2023. ChinaHighPM_2.5: High-resolution and High-quality Ground-level PM_2.5 Dataset for China (2000-2023). [2025-04-10]. https://data.tpdc.ac.cn/zh-hans/data/6168e75d-93ab-4e4a-b7ff-33152e49d0bf (in Chinese).
[43]	Wu, Z.F., Wang, Y., Ren, Y., 2025. Optimizing green space-building landscape characteristics of key urban functional zones for comprehensive thermal environment mitigation. Landscape and Urban Planning. 257, 105274, doi: 10.1016/j.landurbplan.2025.105314.
[44]	Xu, X.L., 2017a. China’s GDP Spatial Distribution Grid Data Set. [2025-04-10]. https://doi.org/10.12078/2017121102 (in Chinese).
[45]	Xu, X.L., 2017b. China’s Population Spatial Distribution Grid Data Set. [2025-04-10]. https://doi.org/10.12078/2017121101 (in Chinese).
[46]	Xue, B., Xiao, X., Li, J.Z., et al., 2023. Multi-source data-driven identification of urban functional areas: A case of Shenyang, China. Chinese Geographical Science. 33(1), 21-35.
[47]	Yang, J., Wang, Y.C., Xiu, C.L., et al., 2020. Optimizing local climate zones to mitigate urban heat island effect in human settlements. Journal of Cleaner Production. 275, 123767, doi: 10.1016/j.jclepro.2020.123767.
[48]	Yang, J., Huang, X., 2021. The 30 m annual land cover dataset and its dynamics in China from 1990 to 2019. Earth System Science Data. 13(8), 3907-3925.
[49]	Yang, J., Yu, W.B., Baklanov, A., et al., 2025. Mainstreaming the local climate zone framework for climate-resilient cities. Nature Communications. 16, 61394, doi: 10.1038/s41467-025-61394-w.
[50]	Yu, W.B., Yang, J., Ren, J.Y., et al., 2025. Identifying localized heat zones in urban heat islands from a hill and saddle perspective. Sustainable Cities and Society. 118, 106028, doi: 10.1016/j.scs.2024.106028.
[51]	Yunnan Provincial Bureau of Statistics, 2020. Yunnan Statistical Yearbook 2020 (No. 36). Beijing: China Statistics Press (in Chinese).
[52]	Zhai, R., Tao, F.L., 2021. Climate change in China affects runoff and terrestrial ecosystem water retention more than changes in leaf area index and land use/cover over the period 1982-2015. Journal of Geophysical Research-Biogeosciences. 126(4), e2020JG005902, doi: 10.1029/2020JG005902.
[53]	Zhang, M.M., Yigit, I., Adiguzel, F., et al., 2024. Impact of urban surfaces on microclimatic conditions and thermal comfort in Burdur, Türkiye. Atmosphere. 15(11), 1375, doi: 10.3390/atmos15111375.
[54]	Zhang, X.L., Kasimu, A., Liang, H.W., et al., 2024a. Mechanism analysis of vegetation phenology in an urban agglomeration in an arid zone driven by seasonal land surface temperatures. Urban Climate. 53, 101278, doi: 10.1016/j.uclim.2024.101278.
[55]	Zhang, X.T., Wang, D.D., Liu, Q., et al., 2019. An operational approach for generating the global land surface downward shortwave radiation product from MODIS data. IEEE Transactions on Geoscience and Remote Sensing. 57(7), 4636-4650.
[56]	Zhao, B.Y., Sun, Z.Q., Wang, M., et al., 2025. Tracking global large lake surface temperature variation from space using MODIS land surface temperature product. Ecological Informatics. 89, 103184, doi: 10.1016/j.ecoint.2025.103184.
[57]	Zhou, L., Hu, F.N., Wang, B., et al., 2022. Relationship between urban landscape structure and land surface temperature: spatial hierarchy and interaction effects. Sustainable Cities and Society. 80, 103795, doi: 10.1016/j.scs.2022.103795.
[58]	Zhu, H.J., Wang, A., Wang, P.T., et al., 2025. Spatiotemporal dynamics and response of land surface temperature and kernel normalized difference vegetation index in Yangtze River Economic Belt, China: Multi-method analysis. Land. 14(3), 598, doi: 10.3390/land14030598.

Data	Spatial resolution	Source
Land surface temperature (LST)	1 km	https://search.earthdata.nasa.gov/
Land cover types (LCTs)	30 m	https://www.ncdc.ac.cn/portal/
Normalized Difference Vegetation Index (DNVI)	1 km	https://search.earthdata.nasa.gov/
Normalized Difference Moisture Index (NDMI)	1 km	https://search.earthdata.nasa.gov/
Digital elevation model (DEM)	30 m	https://www.usgs.gov/
Precipitation (PRE)	1 km	https://data.tpdc.ac.cn/
Downward shortwave radiation (DSR)	1 km	http://www.glass.umd.edu/
Population density (POP)	1 km	http://www.resdc.cn/
Gross domestic product (GDP)	1 km	http://www.resdc.cn/
Particulate matter 2.5 (PM_2.5)	1 km	https://data.tpdc.ac.cn/
Aerosol optical depth (AOD)	1 km	https://modis.gsfc.nasa.gov

Dividing object	Extremely low temperature (°C)	Low temperature (°C)	Medium temperature (°C)	High temperature (°C)	Extremely high temperature (°C)
Annual mean	12.130-9.590	9.590-14.000	14.000-18.410	18.410-22.820	22.820-25.740
Daytime mean	8.760-14.550	14.550-19.040	19.040-23.530	23.530-28.020	28.020-32.910
Nighttime mean	16.390-4.100	4.100-8.780	8.780-13.460	13.460-18.140	18.140-20.280
Spring mean	9.560-12.070	12.070-16.960	16.960-21.805	21.850-26.740	26.740-30.350
Summer mean	0.840-13.720	13.720-17.850	17.850-21.980	21.980-26.110	26.110-30.540
Autumn mean	8.380-10.000	10.000-13.900	13.900-17.800	17.800-21.700	21.700-25.180
Winter mean	-17.170-4.780	4.780-9.520	9.520-14.260	14.260-19.000	19.000-20.350

Year	Training set			Test set
Year	RMSE (°C)	MAE (°C)	R²	RMSE (°C)	MAE (°C)	R²
2005	0.515	0.392	0.987	0.847	0.629	0.966
2010	0.424	0.326	0.991	0.688	0.533	0.977
2015	0.459	0.348	0.989	0.743	0.577	0.970
2020	0.429	0.327	0.991	0.721	0.544	0.975