High-resolution spatial distribution of vegetation biomass and its environmental response on Qinghai-Tibet Plateau: Intensive grid-field survey

Abstract

Due to the complexity of extremely high altitude, topography, and climate, accurately estimating regional biomass is essential yet challenging in alpine regions such as the Qinghai-Tibet Plateau (QTP). Here, we conducted an intensive grid-field survey using a matched-measured biomass dataset of 2,040 field plots on the QTP to obtain high-resolution biomass estimates and their unique response to the environment. The biomass differed significantly among different vegetation types and was highest in evergreen coniferous forests and lowest in alpine grasslands. In addition to traditional mean temperature and precipitation, ultraviolet radiation of growing season (UVGS), and the partial pressure of CO2 (PCO2) were also important factors influencing the spatial variation of vegetation biomass on the QTP. Furthermore, we simulated the spatial distribution of aboveground, belowground, and total biomass on the QTP at a 1-km resolution using the random forest algorithm, with the coefficient of determination (R2) as 0.78, 0.57, and 0.72, respectively. The new estimate of vegetation biomass on the QTP with the intensive field-survey data was 3.16 Gt, including 1.80 Gt in forests, 0.77 Gt in shrubs, 0.52 Gt in grasslands, and 0.07 Gt in deserts. Because of the unique response of biomass to the external environment, such as UVGS and PCO2, we assessed the underlying response of terrestrial ecosystems to global change more comprehensively, especially for sensitive alpine or high-latitude regions in the future. The high-resolution biomass maps may serve as support for regional carbon sink assessment and ecosystem management and conservation, and may provide important parameters for ecological process modeling.

Keywords

• Biomass
• Productivity
• Qinghai-Tibet Plateau
• Spatial variation
• Machine learning
• Adaptation