The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Download
Citation
Articles | Volume XLII-3/W12-2020
https://doi.org/10.5194/isprs-archives-XLII-3-W12-2020-431-2020
https://doi.org/10.5194/isprs-archives-XLII-3-W12-2020-431-2020
06 Nov 2020
 | 06 Nov 2020

NET PRIMARY PRODUCTIVITY AND DRY MATTER IN SOYBEAN CULTIVATION UTILIZING DATAS OF NDVI MULTI-SENSORS

G. Rodigheri, D. C. Fontana, L. P. Schaparini, G. A. Dalmago, and J. Schirmbeck

Keywords: Google Earth Engine, PAR, Remote Sensing, Agriculture

Abstract. Net Primary Productivity (NPP) is an important indicator of vegetation growth status and ecosystems health. NPP can be estimated through remote sensing data, using vegetation indices such as NDVI. However, this index may show systematic differences when using several orbital sensors. Therefore, the objective of this paper was to compare the NDVI data obtained from different sensors and evaluate the impact over the soybean biomass and NPP estimates. NDVI data were recorded from 4 sensors, one on the field and others 3 orbitals sensors (Landsat 8/OLI, Sentinel 2/MSI and Terra/MODIS). Measured data on the field, Photosynthetically Active Radiation (PAR) and Dry Matter (DM), were used to modeling the total DM and also NPP. The NDVI data from different sensors showed differences throughout the cycle, but compared to the reference data there was a correlation greater than 0.84. The DM presented a correlation of 0.91 with the field measured MS data while the NPP presented differences of up to 240 gC/m2/month from in relation to the reference data. Therefore, NDVI obtained from multiple sensors can be used to estimate NPP for surface analysis. However, for more consistent evaluations, a function of adjustment between the NDVI sensor data and NDVI reference data is required, so that the NPP estimation be better correlated to the actual data.