Asynchronous convergence of annual satellite-derived forest productivity and within-year production allocation after fire: Counterfactual evidence from 7197 forest fires in China
6 affiliations
- 1Faculty of Geography; Teaching and Experimentation Base & Innovation Service Center for Geographic Information Technology; GIS Technology Research Center of Resource and Environment in Western China, Ministry of Education
- 2Faculty of Geography; GIS Technology Research Center of Resource and Environment in Western China, Ministry of Education; Corresponding author at: Faculty of Geography, Yunnan Normal University, Kunming 650500, China
- 3Faculty of Surveying and Mapping; Corresponding author at: Faculty of Surveying and Mapping, Kunming Metallurgy University, Kunming 650033, China.
- 4Faculty of Geography
- 5Faculty of Geography; Teaching and Experimentation Base & Innovation Service Center for Geographic Information Technology
- 6Faculty of Geography; GIS Technology Research Center of Resource and Environment in Western China, Ministry of Education
Abstract
Large-scale assessments of post-fire forest recovery commonly evaluate whether annual or growing-season productivity approaches a reference state, yet similar annual totals can arise from different within-year production patterns. For 7197 forest fires in China during 2005–2016, we matched burned forests with contemporaneous unburned controls and reconstructed 8-day counterfactual GPP trajectories through post-fire year 8 using MOD17 and PML-V2.2a. By post-fire years 6–8, annual GPP had approached the contemporaneous unburned reference, with mean relative effects (burned–control differences relative to control GPP) of −1.13% for MOD17 and +0.59% for PML-V2.2a. However, the burned–control relative effect during the early production phase remained 1.98 and 2.45 percentage points lower than during the core phase, respectively. This within-year allocation contrast declined by 77.7% and 73.2% from post-fire year 1 to year 8 but remained negative under four temporal definitions. Control-to-control randomization showed that the national-scale contrast exceeded design-internal variation. Timing differences were concentrated near the leading edge of the production window, and early-phase production amplitude was also lower than core-phase amplitude, although the 8-day data could not reliably separate their contributions. MOD17 and PML-V2.2a reproduced the direction and temporal location of the population-scale pattern, whereas cumulative 8-year GPP differences varied by up to 4.7-fold across MOD17, PML-V2.2a, and GLASS-GPP. These results show that annual satellite-derived productivity and within-year production allocation capture different but complementary dimensions of post-fire recovery. Incorporating within-year temporal structure alongside annual production can improve the characterization of large-scale recovery and the interpretation of multi-product satellite assessments.
Related
- Reducing false alarms in crackmeter-based early warning of small-scale slope deformation via deep learning-driven asynchronous displacement–rainfall data fusion
- Ocean acidification, reduced silicate availability and declines in Ocean productivity
- Growth, biomass productivity, nitrogen uptake and use efficiency of selected enset clones in southern Ethiopia
- Effects of compound microbial fertilizer combined with potassium fulvic acid on soil physicochemical properties, microbial community structure, and cotton productivity in saline-alkali soil
- Optimizing water and nutrient delivery through precision drip fertigation boosts fennel productivity and resource-use efficiency in arid agroecosystems
BibTeX
@article{wang2026asynchronous,
title = {{Asynchronous convergence of annual satellite-derived forest productivity and within-year production allocation after fire: Counterfactual evidence from 7197 forest fires in China}},
author = {Xiao Wang and Quanli Xu and Junhua Yi and Yu Wang and Fengzhi Wu and Shu Wang and Shengfeng Li and Yijing Lin and Zhaoxin Fu and Yajun Wang and Shijie Ye and Ting Yang and Qing Zou},
journal = {Ecological Indicators},
year = {2026},
month = oct,
doi = {10.1016/j.ecolind.2026.115521},
url = {https://doi.org/10.1016/j.ecolind.2026.115521},
note = {CC BY},
}
RIS
TY - JOUR
TI - Asynchronous convergence of annual satellite-derived forest productivity and within-year production allocation after fire: Counterfactual evidence from 7197 forest fires in China
AU - Xiao Wang
AU - Quanli Xu
AU - Junhua Yi
AU - Yu Wang
AU - Fengzhi Wu
AU - Shu Wang
AU - Shengfeng Li
AU - Yijing Lin
AU - Zhaoxin Fu
AU - Yajun Wang
AU - Shijie Ye
AU - Ting Yang
AU - Qing Zou
PY - 2026
DA - 2026/10/23
JO - Ecological Indicators
DO - 10.1016/j.ecolind.2026.115521
UR - https://doi.org/10.1016/j.ecolind.2026.115521
AB - Large-scale assessments of post-fire forest recovery commonly evaluate whether annual or growing-season productivity approaches a reference state, yet similar annual totals can arise from different within-year production patterns. For 7197 forest fires in China during 2005–2016, we matched burned forests with contemporaneous unburned controls and reconstructed 8-day counterfactual GPP trajectories through post-fire year 8 using MOD17 and PML-V2.2a. By post-fire years 6–8, annual GPP had approached the contemporaneous unburned reference, with mean relative effects (burned–control differences relative to control GPP) of −1.13% for MOD17 and +0.59% for PML-V2.2a. However, the burned–control relative effect during the early production phase remained 1.98 and 2.45 percentage points lower than during the core phase, respectively. This within-year allocation contrast declined by 77.7% and 73.2% from post-fire year 1 to year 8 but remained negative under four temporal definitions. Control-to-control randomization showed that the national-scale contrast exceeded design-internal variation. Timing differences were concentrated near the leading edge of the production window, and early-phase production amplitude was also lower than core-phase amplitude, although the 8-day data could not reliably separate their contributions. MOD17 and PML-V2.2a reproduced the direction and temporal location of the population-scale pattern, whereas cumulative 8-year GPP differences varied by up to 4.7-fold across MOD17, PML-V2.2a, and GLASS-GPP. These results show that annual satellite-derived productivity and within-year production allocation capture different but complementary dimensions of post-fire recovery. Incorporating within-year temporal structure alongside annual production can improve the characterization of large-scale recovery and the interpretation of multi-product satellite assessments.
SN - 1470-160X
SN - 1872-7034
ER -