Search

• Calibration and validation of a process-based ecosystem model at an Indian forest. • Modelled and observed annual GPP are 2433 gC m−2 y−1 and 2399 gC m−2 y−1. • Mismatch exists between empirically derived and modelled photosynthetic parameters. • Model simulated sensible and latent heat fluxes concur with direct observations. The ecosystem-atmosphere exchanges of terrestrial ecosystems are key drivers of global carbon, water, and energy cycles which are crucial to be measured accurately and represented in bottom-up biogeochemically and biogeophysically coupled ecosystem, Earth system, and climate models for improving the model predictions and impact assessment of climate change on these ecosystems. The diverse natural ecosystems in India are poorly represented in these models due to the absence any of integrated data-model framework and intercomparison study so far. To partially address this, we have used flux-tower measurements in this study to simulate the gross primary productivity (GPP), sensible (H), and latent heat fluxes (LE) of a moist semi-evergreen deciduous forest in the Kaziranga National Park in subtropical Northeast India using a process-based model ISAM. The model is calibrated using two years of measurement which improves its performance when subsequently run to simulate these fluxes for a year. The model produced annual GPP, mean maximum H, and LE are 2432.26 gC m−2 y−1, 29 and 82 W m−2 respectively as compared to their measured values i.e. 2398.47 gC m−2 y−1, 26 and 73 W m−2 correspondingly. Additionally, we report the calibrated model biogeochemical and biogeophysical parameters which will be useful to simulate the fluxes in this forest using the aforementioned models. [Display omitted] [ABSTRACT FROM AUTHOR]

The terrestrial biosphere has a crucial role in controlling the rate of CO 2 accumulation in the atmosphere. The quantification of Net Ecosystem Exchange (NEE) is critical to determine whether the regional terrestrial ecosystem is a net sink or source of CO 2. This study uses a satellite-data derived light-use-efficiency model (the Vegetation Photosynthesis and Respiration Model, VPRM) to compute the biospheric CO 2 fluxes over India from 2011–2020. A very high-resolution Land Use-Land Cover (LULC) data from the IRS-P6 satellite and MODIS derived surface reflectance are utilised to compute the NEE over distinct vegetation types at ∼ 9 km × 9 km. The VPRM model for the Indian region captures the spatial pattern and seasonal features of NEE over the country. The magnitude of annual mean NEE over India from the VPRM model is relatively high compared to that from Carbon Tracker, FLUXCOM, and Soil Moisture Active Passive Level-4 carbon data products. However, the VPRM model simulated NEE agrees with the NEE observations from the eddy covariance estimates, which are accepted as the ground truth. The fusion of high-resolution LULC and surface reflectance data in VPRM suggests that the average NEE over the Indian region during 2011–2020 is −0.16 ± 0.02 PgC yr−1. Among this, the contribution from Gross Ecosystem Exchange is about −0.47 ± 0.02 PgC yr−1, and Respiration is 0.31 ± 0.01 PgC yr−1. The maximum biospheric CO 2 absorption is during the post-monsoon season, and the minimum is during the pre-monsoon season. The very high spatial resolution and heterogeneity resolved in the NEE data derived in this work offer the first data product of this kind for a variety of NEE analyses over India. [Display omitted] • The Vegetation Photosynthesis and Respiration Model (VPRM) is configured over India. • VPRM computes high-resolution biospheric CO 2 flux over India during 2011-20. • The mean NEE over India during 2011–20 is −0.16 ± 0.02 PgC yr−1. • CO 2 flux optimisation using inverse modelling can benefit from this biospheric flux. [ABSTRACT FROM AUTHOR]