Nonlinear carbon-water coupling in terrestrial ecosystems: Insights from China’s Three-North Shelterbelt Forest region

doi:10.1016/j.regsus.2026.100328

Abstract

Abstract:

Understanding the coupling between carbon and water in terrestrial ecosystems is essential for achieving sustainable development. Net primary productivity (NPP), carbon use efficiency (CUE), and water use efficiency (WUE) are key indicators for assessing carbon balance and carbon-water interactions. However, knowledge gaps remain regarding how these indicators respond to climate change and interact with one another. This study examined the spatial and temporal dynamics of NPP, CUE, and WUE, as well as their interrelationships, within the Three-North Shelterbelt Forest region of China. Furthermore, the study investigated the driving mechanisms of these indicators using Extreme Gradient Boosting (XGBoost), SHapley Additive exPlanations (SHAP), and Partial Least Squares Structural Equation Modeling (PLS-SEM). The results revealed that from 2000 to 2020, both NPP (2.69 g C/(m²•a); P<0.010) and WUE (0.004 g C/(kg H₂O•a); P<0.010) increased significantly, while CUE exhibited a non-significant decline (-5.40×10^-4/a; P>0.050) across different climatic zones (arid, semi-arid, humid, and sub-humid) and vegetation types (cropland, forest, grassland, shrubland, and wetland). The correlation between WUE and NPP (correlation coefficient of 0.70) was stronger than that between CUE and NPP (correlation coefficient of 0.15). NPP and WUE were primarily influenced by leaf area index, whereas CUE was most strongly affected by elevation. The relationships between the key drivers and the three indicators were largely nonlinear, with stronger driver contributions corresponding to more pronounced nonlinear interactions. Moreover, these nonlinear relationships were modulated by differences in dry versus wet climatic conditions. Geographical factors (e.g., longitude, latitude, and elevation) further shaped vegetation characteristics (e.g., fractional vegetation cover and leaf area index) by regulating climatic variables such as temperature, precipitation, and evapotranspiration, ultimately influencing NPP, WUE, and CUE. This study advances the understanding of vegetation carbon-water coupling and provides a scientific basis for ecosystem management and sustainable development policy-making in various climatic zones.

Key words: Carbon use efficiency (CUE), Water use efficiency (WUE), Net primary productivity (NPP), Climate change, Machine learning method, Three-North Shelterbelt Forest

CHEN Xuanhao, LI Chao, ZHANG Shiqiang. Nonlinear carbon-water coupling in terrestrial ecosystems: Insights from China’s Three-North Shelterbelt Forest region[J]. Regional Sustainability, 2026, 7(2): 100328.

Figures/Tables 14

Fig. 1.

Table 1

Table 2

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

Fig. 6.

Fig. 7.

Fig. 8.

Fig. 9.

Fig. 10.

Fig. 11.

Fig. 12.

References 64

[1]	Akaike H., 1974. A new look at the statistical model identification. IEEE Transactions on Automatic Control. 19(6), 716-723. doi: 10.1109/TAC.1974.1100705
[2]	Alsafadi K., Bashir B., Mohammed S., et al., 2024. Response of ecosystem carbon-water fluxes to extreme drought in West Asia. Remote Sensing. 16(7), 1179, doi: 10.3390/rs16071179.
[3]	An X., 2022. Responses of water use efficiency to climate change in evapotranspiration and transpiration ecosystems. Ecological Indicators. 141, 109157, doi: 10.1016/j.ecolind.2022.109157.
[4]	Bian Z.J., Roujean J.F., Fan T.Y., et al., 2023. An angular normalization method for temperature vegetation dryness index (TVDI) in monitoring agricultural drought. Remote Sensing of Environment. 284, 113330, doi: 10.1016/j.rse.2022.113330.
[5]	Chen J.K., Pu J.B., Li J.H., et al., 2024. Response of carbon- and water-use efficiency to climate change and human activities in China. Ecological Indicators. 160, 111829, doi: 10.1016/j.ecolind.2024.111829.
[6]	Chen T.Q., Guestrin C., 2016. XGBoost:A scalable tree boosting system. In: KrishnapuramB., ShahM., SmolaA.J., (Proceedings of the 22^nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.eds.). New York: Association for Computing Machinery, 785-794.
[7]	Cloern J.E., Safran S.M., Smith Vaughn L., et al., 2021. On the human appropriation of wetland primary production. Science of The Total Environment. 785, 147097, doi: 10.1016/j.scitotenv.2021.147097.
[8]	Cui P., Xv D.W., Tang J.N., et al., 2024. Assessing the effects of urban green spaces metrics and spatial structure on LST and carbon sinks in Harbin, a cold region city in China. Sustainable Cities and Society. 113, 105659, doi: 10.1016/j.scs.2024.105659.
[9]	Deng C.L., Zhang B.Q., Cheng L.Y., et al., 2019. Vegetation dynamics and their effects on surface water-energy balance over the Three-North Region of China. Agricultural and Forest Meteorology. 275, 79-90.
[10]	Ding H., Shi X.L., Yuan Z., et al., 2024. Does vegetation greening have a positive effect on global vegetation carbon and water use efficiency? Science of The Total Environment. 951, 175589, doi: 10.1016/j.scitotenv.2024.175589.
[11]	Du L.T., Gong F., Zeng Y.J., et al., 2021. Carbon use efficiency of terrestrial ecosystems in desert/grassland biome transition zone: A case in Ningxia Province, Northwest China. Ecological Indicators. 120, 106971, doi: 10.1016/j.ecolind.2020.106971.
[12]	Du X.Z., Zhao X., Zhou T., et al., 2019. Effects of climate factors and human activities on the ecosystem water use efficiency throughout northern China. Remote Sensing. 11(23), 2766, doi: 10.3390/rs11232766.
[13]	Feng X.M., Fu B.J., Piao S.L., et al., 2016. Revegetation in China’s Loess Plateau is approaching sustainable water resource limits. Nature Climate Change. 6, 1019-1022.
[14]	Fu C.L., Xu M., 2023. Achieving carbon neutrality through ecological carbon sinks: A systems perspective. Green Carbon. 1(1), 43-46. doi: 10.1016/j.greenca.2023.08.005
[15]	Gang C.C., Wang Z.N., You Y.F., et al., 2022. Divergent responses of terrestrial carbon use efficiency to climate variation from 2000 to 2018. Global and Planetary Change. 208, 103709, doi: 10.1016/j.gloplacha.2021.103709.
[16]	Gentine P., Green J.K., Guérin M., et al., 2019. Coupling between the terrestrial carbon and water cycles-a review. Environmental Research Letters. 14(8), 083003, doi: 10.1088/1748-9326/ab22d6.
[17]	Gong H.B., Cao L., Duan Y.F., et al., 2023. Multiple effects of climate changes and human activities on NPP increase in the Three-north Shelter Forest Program area. Forest Ecology and Management. 529, 120732, doi: 10.1016/j.foreco.2022.120732.
[18]	Hair J.F., Matthews L.M., Matthews R.L., et al., 2017. PLS-SEM or CB-SEM: updated guidelines on which method to use. International Journal of Multivariate Data Analysis. 1(2), 107-123. doi: 10.1504/IJMDA.2017.087624
[19]	Hair J.F., Risher J.J., Sarstedt M., et al., 2019. When to use and how to report the results of PLS-SEM. European Business Review. 31(1), 2-24.
[20]	Hou Q.Q., Ji Z.X., Yang H., et al., 2022. Impacts of climate change and human activities on different degraded grassland based on NDVI. Scientific Reports. 12, 15918, doi: 10.1038/s41598-022-19943-6.
[21]	Hou X., Zhang B., He Q.Q., et al., 2024. Spatial-temporal variations in the climate, net ecosystem productivity, and efficiency of water and carbon use in the middle reaches of the Yellow River. Remote Sensing. 16(17), 3312, doi: 10.3390/rs16173312.
[22]	Khalifa M., Elagib N.A., Ribbe L., et al., 2018. Spatio-temporal variations in climate, primary productivity and efficiency of water and carbon use of the land cover types in Sudan and Ethiopia. Science of The Total Environment. 624, 790-806.
[23]	Lam K.L., Liu G., Motelica-Wagenaar A.M., et al., 2022. Toward carbon-neutral water systems: insights from global cities. Engineering. 14, 77-85.
[24]	Lei S., Zhou P., Lin J.Y., et al., 2025. Spatiotemporal variation in carbon and water use efficiency and their influencing variables based on remote sensing data in the Nanling Mountains region. Remote Sensing. 17(4), 648, doi: 10.3390/rs17040648.
[25]	Li B., Huang F., Chang S., et al., 2019a. The variations of satellite-based ecosystem water use and carbon use efficiency and their linkages with climate and human drivers in the Songnen Plain, China. Advances in Meteorology. 1, e8659138, doi: 10.1155/2019/8659138.
[26]	Li B., Huang F., Qin L.J., et al., 2019b. Spatio-temporal variations of carbon use efficiency in natural terrestrial ecosystems and the relationship with climatic factors in the Songnen Plain, China. Remote Sensing. 11(21), 2513, doi: 10.3390/rs11212513.
[27]	Li C., Li X.M., Luo D.L., et al., 2021a. Spatiotemporal pattern of vegetation ecology quality and its response to climate change between 2000-2017 in China. Sustainability. 13(3), 1419, doi: 10.3390/su13031419.
[28]	Li H., Xu F., Li Z.C., et al., 2021b. Forest changes by precipitation zones in northern China after the Three-North Shelterbelt Forest Program in China. Remote Sensing. 13(4), 543, doi: 10.3390/rs13040543.
[29]	Liu C.L., Li W.L., Wang W.Y., et al., 2021. Quantitative spatial analysis of vegetation dynamics and potential driving factors in a typical alpine region on the northeastern Tibetan Plateau using the google earth engine. CATENA. 206, 105500, doi: 10.1016/j.catena.2021.105500.
[30]	Liu X.Y., Liu C., Fan B.H., et al., 2022. Spatial responses of ecosystem water-use efficiency to hydrothermal and vegetative gradients in alpine grassland ecosystem in drylands. Ecological Indicators. 141, 109064, doi: 10.1016/j.ecolind.2022.109064.
[31]	Liu X.Y., Lai Q., Yin S., et al., 2024. Spatio-temporal patterns and control mechanism of the ecosystem carbon use efficiency across the Mongolian Plateau. Science of The Total Environment. 907, 167883, doi: 10.1016/j.scitotenv.2023.167883.
[32]	Lu F., Hu H.F., Sun W.J., et al., 2018. Effects of national ecological restoration projects on carbon sequestration in China from 2001 to 2010. Proceedings of the National Academy of Sciences of the United States of America. 115(16), 4039-4044.
[33]	Lundberg S.M., Lee S.I., 2017. A unified approach to interpreting model predictions. In: GuyonI., LuxburgU.V., BengioS., (Advances in Neural Information Processing Systems 30 (eds.). NIPS 2017). California: NeurIPS Proceedings.
[34]	Luo X.Z., Zhao R.Y., Chu H.S., et al., 2025. Global variation in vegetation carbon use efficiency inferred from eddy covariance observations. Nature Ecology & Evolution. 9, 1414-1425.
[35]	Mu H.W., Li X.C., Wen Y.N., et al., 2022. A global record of annual terrestrial Human Footprint dataset from 2000 to 2018. Scientific Data. 9, 176, doi: 10.1038/s41597-022-01284-8.
[36]	Peng D.L., Wu C.Y., Zhang B., et al., 2016. The influences of drought and land-cover conversion on inter-annual variation of NPP in the three-north shelterbelt program zone of China based on MODIS Data. PLoS ONE. 11(6), e0158173, doi: 10.1371/journal.pone.0158173.
[37]	Potter C.S., Randerson J.T., Field C.B., et al., 1993. Terrestrial ecosystem production: A process model based on global satellite and surface data. Global Biogeochemical Cycles. 7(4), 811-841. doi: 10.1029/93GB02725
[38]	Qin G.X., Meng Z.Y., Fu Y., 2022. Drought and water-use efficiency are dominant environmental factors affecting greenness in the Yellow River Basin, China. Science of The Total Environment. 834, 155479, doi: 10.1016/j.scitotenv.2022.155479.
[39]	Sa R.L., Hua Y.C., Zhai K.T., et al., 2023. Assessing ecological services in the three-north shelter forest area of china using remote sensing. Applied Ecology & Environmental Research. 21(1), 261-285.
[40]	Sanchez G., Trinchera L., Russolillo G., 2013. Plspm: Tools for Partial Least Squares Path Modeling (PLS-PM). [2025-01-26]. https://cran.r-project.org/package=plspm.
[41]	Shao H.B., Chu L.Y., Jaleel C.A., et al., 2009. Understanding water deficit stress-induced changes in the basic metabolism of higher plants - biotechnologically and sustainably improving agriculture and the ecoenvironment in arid regions of the globe. Critical Reviews in Biotechnology. 29(2), 131-151. doi: 10.1080/07388550902869792
[42]	Shao R., Zhang B.Q., Su T.X., et al., 2019. Estimating the increase in regional evaporative water consumption as a result of vegetation restoration over the Loess Plateau, China. Journal of Geophysical Research-Atmospheres. 124(22), 11783-11802. doi: 10.1029/2019JD031295
[43]	Sippel S., Reichstein M., Ma X., et al., 2018. Drought, heat, and the carbon cycle: a review. Current Climate Change Reports. 4, 266-286.
[44]	Tang X.L., Carvalhais N., Moura C., et al., 2019. Global variability of carbon use efficiency in terrestrial ecosystems. Biogeosciences Discussions. doi: 10.5194/bg-2019-37.
[45]	Tarin T., Nolan R.H., Eamus D., et al., 2020. Carbon and water fluxes in two adjacent Australian semi-arid ecosystems. Agricultural and Forest Meteorology. 281, 107853, doi: 10.1016/j.agrformet.2019.107853.
[46]	Tian H.Q., Chen G.S., Liu M.L., et al., 2010. Model estimates of net primary productivity, evapotranspiration, and water use efficiency in the terrestrial ecosystems of the southern United States during 1895-2007. Forest Ecology and Management. 259(7), 1311-1327. doi: 10.1016/j.foreco.2009.10.009
[47]	Wang L.X., Gao J.X., Zhang W.G., et al., 2022. Carbon Sequestration in Vegetation and its Change in the Three-North Shelter Forest Region of China in 2000-2021. [2025-02-22]. https://www.researchsquare.com/article/rs-1710152/v1.
[48]	Wang Y., Lü Y.H., Lü D., et al., 2024a. Carbon and water relationships change nonlinearly along elevation gradient in the Qinghai Tibet Plateau. Journal of Hydrology. 628, 130529, doi: 10.1016/j.jhydrol.2023.130529.
[49]	Wang Y.P., Zhang L., Liang X., et al., 2024b. Coupled models of water and carbon cycles from leaf to global: A retrospective and a prospective. Agricultural and Forest Meteorology. 358, 110229, doi: 10.1016/j.agrformet.2024.110229.
[50]	Wang Z.Y., Zhou Y.F., Sun X.Y., et al., 2024c. Estimation of NPP in Huangshan District based on deep learning and CASA model. Forests. 15(8), 1467, doi: 10.3390/f15081467.
[51]	Wu X.M., Zhou T., Zeng J.Y., et al., 2025. Application of a Random Forest method to estimate the water use efficiency on the Qinghai Tibetan Plateau during the 1982-2018 growing season. Remote Sensing. 17(3), 527, doi: 10.3390/rs17030527.
[52]	Xia L., Wang F., Mu X.M., et al., 2015. Water use efficiency of net primary production in global terrestrial ecosystems. Journal of Earth System Science. 124(5), 921-931. doi: 10.1007/s12040-015-0587-4
[53]	Xiao B.Q., Bai X.Y., Zhao C.W., et al., 2023. Responses of carbon and water use efficiencies to climate and land use changes in China’s karst areas. Journal of Hydrology. 617, 128968, doi: 10.1016/j.jhydrol.2022.128968.
[54]	Xie X.H., Liang S.L., Yao Y.J., et al., 2015. Detection and attribution of changes in hydrological cycle over the Three-North region of China: Climate change versus afforestation effect. Agricultural and Forest Meteorology. 203, 74-87.
[55]	Xu H.J., Wang X.P., Yang T.B., 2017. Trend shifts in satellite-derived vegetation growth in Central Eurasia, 1982-2013. Science of The Total Environment. 579, 1658-1674.
[56]	Xu H.J., Wang X.P., Zhao C.Y., 2021. Drought sensitivity of vegetation photosynthesis along the aridity gradient in northern China. International Journal of Applied Earth Observation and Geoinformation. 102, 102418, doi: 10.1016/j.jag.2021.102418.
[57]	Xu Z., Tian Y., Liu Z.W., et al., 2023. Comprehensive effects of atmosphere and soil drying on stomatal behavior of different plant types. Water. 15(9), 1675, doi: 10.3390/w15091675.
[58]	Yan Y.C., Liu X.P., Ou J.P., et al., 2018. Assimilating multi-source remotely sensed data into a light use efficiency model for net primary productivity estimation. International Journal of Applied Earth Observation and Geoinformation. 72, 11-25.
[59]	Yang J., Zhang X.C., Luo Z.H., et al., 2017. Nonlinear variations of net primary productivity and its relationship with climate and vegetation phenology, China. Forests. 8(10), 361, doi: 10.3390/f8100361.
[60]	Yin J.B., Gentine P., Slater L., et al., 2023. Future socio-ecosystem productivity threatened by compound drought-heatwave events. Nature Sustainability. 6, 259-272.
[61]	Yuan F.H., Liu J.Z., Zuo Y.J., et al., 2020. Rising vegetation activity dominates growing water use efficiency in the Asian permafrost region from 1900 to 2100. Science of The Total Environment. 736, 139587, doi: 10.1016/j.scitotenv.2020.139587.
[62]	Zhang M., Wang J.M., Li S.J., 2019. Tempo-spatial changes and main anthropogenic influence factors of vegetation fractional coverage in a large-scale opencast coal mine area from 1992 to 2015. Journal of Cleaner Production. 232, 940-952.
[63]	Zhang Q.A., Chen W., 2021. Ecosystem water use efficiency in the Three-North Region of China based on long-term satellite data. Sustainability. 13(14), 7977, doi: 10.3390/su13147977.
[64]	Zhu W.Q., Pan Y.Z., He H., et al., 2006. Simulation of maximum light use efficiency for some typical vegetation types in China. Chinese Science Bulletin. 51, 457-463.

Variable	Data product	Spatial resolution	Temporal resolution	Source
LST	MOD11A2	1.0 km	8 d	https://search.earthdata.nasa.gov
NDVI	MOD13A1	0.5 km	16 d	https://search.earthdata.nasa.gov
LAI	MOD15A2H	0.5 km	8 d	https://search.earthdata.nasa.gov
ET	MOD16A2	0.5 km	8 d	https://search.earthdata.nasa.gov
GPP	MOD17A2H	0.5 km	8 d	https://search.earthdata.nasa.gov
GPP	MOD17A3HGF	0.5 km	1 a	https://search.earthdata.nasa.gov
NPP	MOD17A3HGF	0.5 km	1 a	https://search.earthdata.nasa.gov
VCT	MCD12Q1	0.5 km	1 a	https://search.earthdata.nasa.gov
TEM	Climatic data	0.0083^o	1 month	http://data.tpdc.ac.cn/
PRE	Climatic data	0.0083^o	1 month	http://data.tpdc.ac.cn/
PET	Climatic data	0.0083^o	1 month	http://data.tpdc.ac.cn/
SNR	ERA5	0.1000^o	1 month	https://cds.climate.copernicus.eu/
VPD	Terra climatology	4.0 km	1 month	https://www.climatologylab.org/
CO₂	Mauna Loa CO₂	1.0000^o	1 a	https://www.esrl.noaa.gov/gmd/ccgg/trends/data.html
HF	Human footprint	1.0 km	1 a	https://figshare.com/articles/figure/An_annual_global_terrestrial_Human_Footprint_dataset_from_2000_to_2018/16571064

Vegetation type	ε_max (g C/MJ)	SNR_max (MJ/m²)	Vegetation type	ε_max (g C/MJ)	SNR_max (MJ/m²)
Forest	0.485	5.170	Grassland	0.542	4.460
Cropland	0.542	4.460	Wetland	0.542	4.460
Shrubland	0.429	4.490	Bare ground	0.542	4.460