Spatiotemporal estimation of gross primary production for terrestrial wetlands using satellite and field data

• Published in: Remote sensing applications 2022-08, Vol.27, p.100786, Article 100786
• Main Authors: Dąbrowska-Zielińska, Katarzyna, Misiura, Katarzyna, Malińska, Alicja, Gurdak, Radosław, Grzybowski, Patryk, Bartold, Maciej, Kluczek, Marcin
• Format: Article
• Language: English
• Subjects: air temperature; carbon; Chamber measurements; evapotranspiration; Gross primary production; gross primary productivity; latent heat; Modelling; net ecosystem exchange; peatlands; phenology; photosynthesis; photosynthetically active radiation; remote sensing; Satellite data; satellites; surface temperature; variance; vegetation; Wetlands
• ISSN: 2352-9385; 2352-9385
• DOI: 10.1016/j.rsase.2022.100786

The main goal of this study was to determine the GPP at the Biebrza Wetlands for sedges, reeds and grasses habitats, applying satellite data. The ground measurements of Net Ecosystem Exchange (NEE) and respiration (RESP) were conducted at 26 sites at the peatland using chamber method during the vegetation growth from April to October 2015–2020. Applying round-the-clock registrations of Photosynthetically Active Radiation (PAR) and one ground measurement of NEE and RESP the simulation of the daily course of GPP was performed. GPP daily mean values were the object of estimation applying the models based on the satellite data derived from Sentinel 2, Sentinel 3 and MODIS. The main factors of photosynthetic carbon uptake – greenness and moisture were described by satellite vegetation indices Normalized Difference Vegetation Index (NDVI), Normalized Difference Infrared Index (NDII), Accumulated Photosynthetically Active Radiation (APAR), evapotranspiration, latent heat and the differences of surface temperature (Ts) and air temperature (Ta). The models presented as a result of this research can be used applying solely the satellite data. The results of the analysis revealed that NDVI index is the driving factor for the assessment of GPP independently on the type of satellite data (r = 0.64) Up to 40% of GPP variance is explained by the variable connected with the vegetation greenness. The contribution of vegetation moisture indices is about two times smaller. The best fit has been achieved for the Sentinel-3 models with inclusion of the environmental parameters, as NDVI, temperature difference (Ts–Ta), latent heat. The differences of uptake the CO2 between the habitats depending on their phases of phenological development have been presented.