Preprints
https://doi.org/10.5194/egusphere-2026-6664
https://doi.org/10.5194/egusphere-2026-6664
22 Nov 2026
 | 22 Nov 2026
Status: this preprint is open for discussion and under review for Geoscientific Model Development (GMD).

VISIT-4type v1.0: Forest age structure, wood harvest and land-cover transitions in a dynamic global vegetation model

Jianyong Dong, Shuting Zhao, Ximing Chen, Zhangcai Yin, and Akihiko Ito

Abstract. In dynamic global vegetation models (DGVMs), representing land-use change solely through net changes in land-cover fractions can obscure forest disturbance histories and associated carbon transfers. Here we describe VISIT-4type (v1.0), which extends the two-type framework of the Vegetation Integrative SImulator for Trace gases (VISIT) to primary forest, secondary forest, cropland and other natural vegetation. Both forest types are resolved into six age cohorts; a transition ledger transfers carbon with changing area; wood harvest follows Land-Use Harmonization 2 (LUH2) area requests with separate fuelwood and industrial-wood pathways; and secondary-forest leaf traits are derived from the TRY database. Six model configurations under common environmental forcing are compared to quantify the contributions of these developments to differences in simulated net biome productivity (NBP) over 1850–2019. Against eddy-covariance observations, VISIT-4type reduced the per-site root-mean-square error (RMSE) of annual gross primary production (GPP), net ecosystem exchange (NEE) and ecosystem respiration at 64 %, 74 % and 74 % of forest sites (p ≤ 0.015); the median RMSE decreased from 615 to 565, 303 to 286 and 499 to 434 g C m⁻² yr⁻¹, without significant changes in correlation. Against FLUXCOM X-BASE (2001–2019), the area-weighted mean RMSE of monthly GPP and NEE decreased by 0.42 and 0.25 g C m⁻² month⁻¹, with lower errors over 46.7 % and 56.1 % of the evaluated land, concentrated in grid cells with a history of wood harvest or forest clearing. Harvested stem carbon was 44–61 % of LUH2 wood-harvest biomass, reflecting lower simulated carbon density in harvested primary forest and in the young secondary forests that dominate the harvest requests. Over 1850–2019, global mean NBP was 0.173 Pg C yr⁻¹ lower in VISIT-4type than in VISIT-2type. Wood harvest with forest age structure accounted for −0.171 Pg C yr⁻¹ of this difference, whereas land-cover structure and transitions contributed +0.026 Pg C yr⁻¹ and secondary-forest traits −0.027 Pg C yr⁻¹. Relative to 1850, the 2019 ecosystem carbon-stock difference between the two versions (−81.2 Pg C) was about 52 Pg C more negative than the cumulative NBP difference over 1851–2019 (−29.2 Pg C). About 45 Pg C of this gap arises because abandoned cropland in VISIT-2type inherits the per-area carbon of natural vegetation without a corresponding uptake flux, which are not included in its ecosystem stock. These results show that disturbed-stand carbon density and the end use of harvested wood are key controls on simulated land-use effects, and that flux evaluation needs to be complemented by forest observations that resolve stand age and successional stage.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Share
Jianyong Dong, Shuting Zhao, Ximing Chen, Zhangcai Yin, and Akihiko Ito

Status: open (until 17 Jan 2027)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
Jianyong Dong, Shuting Zhao, Ximing Chen, Zhangcai Yin, and Akihiko Ito
Jianyong Dong, Shuting Zhao, Ximing Chen, Zhangcai Yin, and Akihiko Ito
Metrics will be available soon.
Latest update: 22 Nov 2026
Download
Short summary
Computer models of the land carbon cycle usually track how much land is forest or farmland, but not how old the forests are or what happens to harvested wood. We extended a global vegetation model to separate old-growth from regrowing forests, follow forest age, and route harvested wood to fuel, products and residues. The new model matched measured carbon exchange better and showed that the carbon density of cut forests and the use of the wood strongly shape estimates of land carbon uptake.
Share