Spatiotemporal prediction of forest litterfall in China by using multi-source data and Transformer-CatBoost model

Menglei Guo , Huaiqing Zhang , Jingwei Tan , Yang Liu , Sihan Chen , Hao Lei , Yukai Shi

Journal of Forestry Research ›› 2025, Vol. 37 ›› Issue (1) : 24

PDF
Journal of Forestry Research ›› 2025, Vol. 37 ›› Issue (1) :24 DOI: 10.1007/s11676-025-01962-4
Original Paper
research-article

Spatiotemporal prediction of forest litterfall in China by using multi-source data and Transformer-CatBoost model

Author information +
History +
PDF

Abstract

Forest litterfall is a key contributor to soil carbon accumulation. However, existing studies have primarily foused on site-level observations or annual-scale assessments, while the intra-annual dynamics and spatial distribution of forest litterfall at the national scale remain poorly understood. In turn, this limitied comprehensive spatiotemporal assessments of forest carbon sequestration capacity. In this study, we compiled 4,223 monthly litterfall observations from 88 forest sites across China and integrated multi-source environmental variables to develop a Transformer-CatBoost hybrid prediction model for estimating the spatiotemporal patterns of forest litterfall across three representatibe years (2002, 2009 and 2018), corresponding to major stages of ecological restoration efforts in China. Model evaluation demonstrated strong predictive performance (R2 = 0.74), effectively capturing the nonlinear relationships driving litterfall dynamics. By incorporating national forest area changes in 2002, 2009, and 2018, the study further revealed the spatiotemporal evolution of forest structure under large-scale ecological restoration programs. Based on nationwide monthly-scale modeling results, we systematically characterized the spatial distribution and seasonal variation of litterfall production across China’s forests, with an anuual average of 547.04 ± 0.23 g m⁻2 (or 479.13 ± 0.20 g m⁻2 excluding January and December). Furthermore, using a fixed carbon conversion rate, we estimated national carbon content of forest litterfall at 290.4 Tg in 2002, 311.9 Tg in 2009, and 354.1 Tg in 2018, indicating a clear increasing trend. This study represents the nationwide, monthly-scale modeling and prediction of forest litterfall in China.

Graphical Abstract

Keywords

Forest litterfall / Carbon sequestration / Spatiotemporal prediction / Forest ecosystem / Transformer-CatBoost

Cite this article

Download citation ▾
Menglei Guo, Huaiqing Zhang, Jingwei Tan, Yang Liu, Sihan Chen, Hao Lei, Yukai Shi. Spatiotemporal prediction of forest litterfall in China by using multi-source data and Transformer-CatBoost model. Journal of Forestry Research, 2025, 37(1): 24 DOI:10.1007/s11676-025-01962-4

登录浏览全文

4963

注册一个新账户 忘记密码

References

[1]

Aguirre-Gutiérrez J, Rifai S, Deng XJ, Ter Steege H, Thomson E, Corral-Rivas J, Guimaraes A, Muller S, Klipel J, Fauset S. Canopy functional trait variation across Earth’s tropical forests. Nature, 2025, 641: 129-136.

[2]

Angst G, Pokorný J, Mueller CW, Prater I, Preusser S, Kandeler E, Meador T, Straková P, Hájek T, van Buiten G. Soil texture affects the coupling of litter decomposition and soil organic matter formation. Soil Biol Biochem, 2021, 159: 108302

[3]

Baietto A, Hirigoyen A, Hernández J, del Pino A. Litterfall production modeling based on climatic variables and nutrient return from stands of Eucalyptus grandis Hill ex Maiden and Pinus taeda L. J Forestry Res, 2024, 35(1): 61

[4]

Beugnon R, Albert G, Hähn G, Yu W, Haider S, Hättenschwiler S, Davrinche A, Rosenbaum B, Gauzens B, Eisenhauer N. Improving forest ecosystem functions by optimizing tree species spatial arrangement. Nat Commun, 2025, 16(1): 6286

[5]

Chang L, Deng J, Zhang J, Fu Q, Wang T, Osono T, Peng H, Robson TM, Kurokawa H, Wang Q-W. Sunlight promotes aboveground carbon loss by producing polysaccharides from litter decomposition in a temperate forest. J Forestry Res, 2025, 36(1): 22

[6]

Cheng K, Yang HT, Guan HC, Ren Y, Chen YL, Chen MX, Yang ZK, Lin DY, Liu WY, Xu JC. Unveiling China’s natural and planted forest spatial–temporal dynamics from 1990 to 2020. ISPRS J Photogramm Remote Sens, 2024, 209: 37-50.

[7]

Domeignoz-Horta LA, Cappelli SL, Shrestha R, Gerin S, Lohila AK, Heinonsalo J, Nelson DB, Kahmen A, Duan P, Sebag D. Plant diversity drives positive microbial associations in the rhizosphere enhancing carbon use efficiency in agricultural soils. Nat Commun, 2024, 15(1): 8065

[8]

Domke GM, Perry CH, Walters BF, Woodall CW, Russell MB, Smith JE. Estimating litter carbon stocks on forest land in the United States. Sci Total Environ, 2016, 557-558: 469-478.

[9]

Feng YY, Yang MY, Chen H, Zhang K, Ran FJ, Chen ZY, Yang HJ. Synergistic effects of environmental factors on benthic diversity: machine learning analysis. Water Res, 2025, 282: 123789

[10]

Gao W, Huang MG, Huang YR, Wu XS, Fang DL, Chen AP, Huang SD. Dynamic characteristics of litterfall and carbon and nitrogen return in three forest types in subtropical China. Res Soil Water Conserv, 2023, 30(04): 146-153.

[11]

Girardin CAJ, Malhi Y, Aragao L, Mamani M, Huaraca Huasco W, Durand L, Feeley KJ, Rapp J, Silva‐Espejo JE, Silman M. Net primary productivity allocation and cycling of carbon along a tropical forest elevational transect in the Peruvian Andes. Glob Change Biol, 2010, 16(12): 3176-3192.

[12]

Guo YL, Li DX, Wang B, He YL, Xiang WS, Jiang YL, Li XK. Composition and spatio-temporal dynamics of litter fall in a northern tropical karst seasonal rainforest in Nonggang, Guangxi, southern China. Biodivers Sci, 2017, 25(03): 265-274.

[13]

Guo XW, Zhang YX, You YM, Sun JX. Research advance in the effects of litter input on forest soil organic carbon transformation and stability. Chin J Appl Ecol, 2024, 35(09): 2352-2361.

[14]

He Y, Wang XH, Wang K, Tang SC, Xu H, Chen AP, Ciais P, Li XY, Peñuelas J, Piao SL. Data-driven estimates of global litter production imply slower vegetation carbon turnover. Glob Change Biol, 2021, 27(8): 1678-1688.

[15]

Hu Y, Yang Y, Wei Y, Li XZ, Jiao Y, Liao JP, Fu RY, Dai LC, Hu ZM. Contrasting spatial variations between above and below-ground net primary productivity in global grasslands. Ecol Indic, 2025, 170: 113083.

[16]

Huang SD, Huang YR, Gao W, Nie S, Cai B, Lin J. Dynamics of litterfall and nutrient return in three typical forests of Wuyi Mountain along altitudinal gradient. J Trop Subtrop Bot, 2020, 28(04): 394-402

[17]

Jiang JJ, Feng LX, Hu JG, Zhu C. Interpretable machine learning unveils threshold responses and spatial patterns of global soil respiration. Ecol Indic, 2025, 177: 113750

[18]

Lai ST, Dai WT, Wu FZ, Zhu B, Ni XY. Labile carbon release from plant litter and its effect on soil organic matter formation in a subtropical forest. Plant Soil, 2024, 511: 151-161.

[19]

Lan GF, Zhang Q, Chen XB, Chen SD, Xiong DC, Liu XF, Yang ZJ, Yang YS. Seasonal dynamics of litterfall of a Castanopsis kawakamii evergreen broadleaf forest in mid-subtropical China. Chin J Plant Ecol, 2024, 48(12): 1589

[20]

Li SH, Yuan WP, Ciais P, Viovy N, Ito A, Jia BR, Zhu D. Benchmark estimates for aboveground litterfall data derived from ecosystem models. Environ Res Lett, 2019, 14(8): 084020.

[21]

Li K, Zhao WD, Zhu CS, Lin X, Lin Y, He ZM. Monthly litter production of Acacia aulacocarpa and Pinus elliottii artificial forest and its relationship with meteorological factors. J for Environ, 2021, 41(06): 570-575.

[22]

Li H, Cao Y, Wu Y, Liu S, Zhao W, Zhou G, Xiao J, Alexandrov G, Qiu L. Data-driven modeling indicates projected increase in plant production confines warming-induced topsoil organic carbon change in China within a small range in the 21st century. Sustain Horiz, 2025, 15: 100138.

[23]

Li Q, Yin J, Zhang XX, Hao DL, Ferreira MP, Yan WH, Tian Y, Zhang D, Tan S, Nie S. Tree-level carbon stock estimations across diverse species using multi-source remote sensing integration. Comput Electron Agric, 2025, 231: 109904.

[24]

Lin WQ, Gao W, Ye GF, Huang SD, Huang YR, Yue XJ. Litter nutrient return of different plantations in a coastal sand dune of southern subtropical region. J for Environ, 2019, 39(03): 225-231.

[25]

Liu X, Zhou T, Luo H, Xu PP, Gao S, Liu JJ. Models ignoring spatial heterogeneities of forest age will significantly overestimate the climate effects on litterfall in China. Sci Total Environ, 2019, 661: 492-503.

[26]

Liu XD, Feng YJ, Zhao XY, Cui ZJ, Liu PL, Chen XZ, Zhang QM, Liu JX. Climatic drivers of litterfall production and its components in two subtropical forests in South China: a 14-year observation. Agric for Meteorol, 2024, 344: 109798.

[27]

Lyu M, Homyak PM, Xie JS, Peñuelas J, Ryan MG, Xiong XL, Sardans J, Lin WS, Wang MH, Chen GS. Litter quality controls tradeoffs in soil carbon decomposition and replenishment in a subtropical forest. J Ecol, 2023, 111(10): 2181-2193.

[28]

Malhi Y, Doughty C, Galbraith D. The allocation of ecosystem net primary productivity in tropical forests. Philos Trans R Soc Lond B Biol Sci, 2011, 366(1582): 3225-3245.

[29]

Neumann M, Ukonmaanaho L, Johnson J, Benham S, Vesterdal L, Novotný R, Verstraeten A, Lundin L, Thimonier A, Michopoulos P. Quantifying carbon and nutrient input from litterfall in European forests using field observations and modeling. Glob Biogeochem Cycles, 2018, 32(5): 784-798.

[30]

Pfeffer MA, Ling SSH, Wong JKW. Exploring the frontier: Transformer-based models in EEG signal analysis for brain-computer interfaces. Comput Biol Med, 2024, 178: 108705.

[31]

Pietrzykowski M, Świątek B, Woś B, Klamerus-Iwan A, Mąsior P, Pająk M, Gruba P, Likus-Cieślik J, Tabor J, Ksepko M. The effect of forest disturbances and regeneration scenario on soil organic carbon pools and fluxes: a review. J Forestry Res, 2024, 36(1): 12.

[32]

Prescott CE, Vesterdal L. Decomposition and transformations along the continuum from litter to soil organic matter in forest soils. For Ecol Manage, 2021, 498: 119522.

[33]

Qian Z, Xu X, Wang T, Yang C. Dynamic study on the amount of forest litter in secondary forests of Kuankuoshui Nature Reserve. Contemp Hortic, 2020, 43(19): 51-52+101

[34]

Qian ZM, He L, Dou QL, Wu XX, Li JX. Dynamics of litterfall in Kuankuoshui Chinese fir plantation. Journal of Zunyi Normal University, 2020, 22(04): 81-84

[35]

Rashid H, Sun XY, Wu FZ, Ni XY. Nutrient leaching after subtropical forest conversions and its implications for soil fertility. For Ecol Manage, 2024, 566: 122044.

[36]

Roeder M, Dossa GG, Cornelissen JHC, Yang X, Nakamura A, Tomlinson KW. Liana litter decomposes faster than tree litter in a multispecies and multisite experiment. J Ecol, 2022, 110(10): 2433-2447.

[37]

Shen GR, Xiang QQ, Chen DM, Wu Y, Liu CJ. Spatio-temporal distribution characteristics of forest litterfall in China. Chin J Appl Ecol, 2017, 28(08): 2452-2460.

[38]

Shi JZ, Xu H, Lin MX, Li YD. Dynamics of litterfall production in the tropical mountain rainforest of Jianfengling,Hainan Island,China. Plant Sci J, 2019, 37(05): 593-601

[39]

Sun XF, Liu F, Zhang QZ, Li YC, Zhang LF, Wang J, Zhang HY, Wang CK, Wang XC. Biotic and climatic controls on the interannual variation in canopy litterfall of a deciduous broad-leaved forest. Agric for Meteorol, 2021, 307: 108483.

[40]

Tan YB, Zheng W, He F, Lu GD. Litter compositions and time dynamics in natural forests of Castanopsis hystrix, in Rongxian County, Guangxi Province. J Cent South Univ for Technol, 2015, 35(06): 7-10.

[41]

Tan FY, Han JX, Zhang YF, Xie WD, Liang JL, Lu LG, Cao YH. Dynamic changes of litterfall and its decomposition of three forest types in coastal sandy land. Southwest China J Agric Sci, 2023, 36(07): 1513-1521.

[42]

Teng QM, Fang T, Zhang QQ, Gunina A, Zheng AY, Song ZL, Zhou JY, Chang SX, Li YC. Successional transition from broadleaf to bamboo forests promotes fungal communities and soil carbon mineralization following the altered litterfall quality. Appl Soil Ecol, 2025, 209: 106006.

[43]

Wang L, Lu WL, Song Y, Liu SJ, Fu YV. Using machine learning to identify environmental factors that collectively determine microbial community structure of activated sludge. Environ Res, 2024, 260: 119635

[44]

Wang YX, Wu HX, Dong JX, Qin G, Zhang HR, Liu Y, Qiu YZ, Wang JM, Long MS. Timexer: Empowering transformers for time series forecasting with exogenous variables. Adv Neural Inf Process Syst, 2024, 37: 469-498. DOI:

[45]

Wang S, Han W, Yi S. Afforestation species and slope as key drivers of soil carbon sequestration in plantations of the tropical-subtropical transition zone: a case study from Xishuangbanna, Southwest China. J for Res, 2025, 36(1): 1-14.

[46]

Wang XY, Chen WB, Niu JZ, Lun XX. Analysis of soil organic carbon characteristics and carbon sequestration potential of different forest types. Sci Soil Water Conserv, 2025, 23(04): 175-183.

[47]

Wang Y, Liu CN, Liu BY, Lee TM. Integrated outcomes of large-scale ecological restoration projects on biodiversity–eco-environment–society in China. Geogr Sustain, 2025, 6(3): 100243.

[48]

Wu QQ, Wang CK, Zhang QZ. Inter- and intra-annual dynamics in litter production for six temperate forests. Acta Ecol Sin, 2017, 37(03): 760-769

[49]

Wu XJ, Shi Y, Zhu JN, Sun LJ, Ma XJ. Impacts of global warming on forest litters. World for Res, 2023, 36(01): 26-32.

[50]

Wu QX, Ni XY, Sun XY, Chen ZH, Hong SB, Berg B, Zheng MH, Chen J, Zhu JJ, Ai L. Substrate and climate determine terrestrial litter decomposition. Proc Natl Acad Sci U S A, 2025, 122(7): e2420664122

[51]

Wu X, Ouyang SN, Tan XP, Bose AK, Cheng W, Tie LH. Effects of understory vegetation and climate change on forest litter decomposition: implications for plant and soil management. Plant Soil, 2025, 512: 1-25.

[52]

Xie LL, Li Y, Zhang ZY, Siddique KH, Song XY. Exploring the combined effects of drought and drought-flood abrupt alternation on vegetation using interpretable machine learning model and r-vine copula function. Agric for Meteorol, 2025, 370: 110568.

[53]

Yang X, Szlavecz K, Pitz SL, Langley JA, Chang CH. The partitioning of litter carbon fates during decomposition under different rainfall patterns: a laboratory study. Biogeochemistry, 2020, 148(2): 153-168.

[54]

You CM, Wu FZ, Yang WQ, Tan B, Yue K, Ni XY. The National Key Forestry Ecology Project has changed the zonal pattern of forest litter production in China. For Ecol Manag, 2017, 399: 37-46.

[55]

Yun LL, Zhang HD, Yi D and Huang X (2024) Dynamics and decomposition characteristics of litter of two typical forest species. Liaoning Forestry Science and Technology. (03): 11–16+68

[56]

Zhai JJ, Wang L, Liu Y, Wang CY, Mao XG. Assessing the effects of China’s three-north shelter forest program over 40 years. Sci Total Environ, 2023, 857: 159354

[57]

Zhang DQ, Hui DF, Luo YQ, Zhou GY. Rates of litter decomposition in terrestrial ecosystems: global patterns and controlling factors. J Plant Ecol, 2008, 1(2): 85-93.

[58]

Zhang XP, Wang XP, Zhu B, Zong ZJ, Peng CH, Fang JY. Litter fall production in relation to environmental FAC-TORS in Northeast China’s forests. Chin J Plant Ecol, 2008, 05: 1031-1040

[59]

Zhang LM, Wang CK, Tang Y. Changes in structural components and respiration rates of coarse woody debris at the initial decomposition stage for 11 temperate tree species. Acta Ecol Sin, 2011, 31(17): 5017-5024.

[60]

Zhang YX, Hu WF, Luo M, Tong C, Zhao LL, Chen KL, Huang JF. Litter production and decomposition dynamics of Kandelia obovata from a mangrove swamp in Min River estuary. Acta Sci Circumstantiae, 2019, 39(04): 1312-1323.

[61]

Zhang WP, Fornara D, Yang H, Yu RP, Callaway RM, Li L. Plant litter strengthens positive biodiversity–ecosystem functioning relationships over time. Trends Ecol Evol, 2023, 38(5): 473-484.

[62]

Zhang LL, Yadav V, Zhu HC, Shi YY, Hu XX, Wang XX, Zhou XP, Subhash B, Kang YX. Climate drivers of litterfall biomass dynamics in three types of forest stands on the Loess Plateau. Ecol Indic, 2024, 163: 112088.

[63]

Zhang HQ, Zhang T, Zhou YD, Huang CX, Qiu XC, Feng G, Long WX, Ding Y. Litterfall production in tropical natural forests on Hainan Island: variations across vegetation types and dependence on community properties. Ecol Indic, 2025, 177: 113759.

[64]

Zhang MN, Liu SR, Luo XZ, Keenan TF, Fu LY, Xiao CW, Zhang Y, Gong P. Incorporating site suitability and carbon sequestration of tree species into China’s climate-adaptive forestation. Sci Bull, 2025, 70(11): 1834-1845.

[65]

Zhao XL, Tang XL, Du J, Pei XJ, Chen G, Xu TT. A data-driven estimate of litterfall and forest carbon turnover and the drivers of their inter-annual variabilities in forest ecosystems across China. Sci Total Environ, 2022, 821: 153341

[66]

Zhao HM, Hu GQ, Hu YT, Cheng JH, Yang WJ. Effects of plant life-form and diversity on litter decomposition in a temperate desert. Acta Ecol Sin, 2025, 14: 1-13.

[67]

Zhao YD, Lu N, Shi H, Huang JB, Fu BJ. Patterns and driving factors of litter decomposition rates in global dryland ecosystems. Glob Change Biol, 2025, 31(1): e70025

[68]

Zhou SQ, Ai L, Ni XY, Wu FZ, Wu QX, Zhu JJ, Zhang XY. Global patterns and controls of variation in cellulose decomposition rates of plant litters. Chin J Plant Ecol, 2025, 49(03): 393-403.

[69]

Zhu MH, Fanin N, Kui WQ, Chao XZ, Liang S, Ye J, Lin F, Yuan ZQ, Mao ZK, Wang XG, Hao ZQ. High functional breadth of microbial communities decreases home-field advantage of litter decomposition. Soil Biol Biochem, 2024, 188: 109232

[70]

Zou WT, Chen SZ, Fang Q, Zhang C. Research advances in China’s forest quality evaluation indicators. World for Res, 2024, 37(05): 67-72.

RIGHTS & PERMISSIONS

The Author(s)

PDF

29

Accesses

0

Citation

Detail

Sections
Recommended

/