xCO2 temporal variability above Brazilian agroecosystems: A remote sensing approach

Abstract

Agriculture and soil management practices are closely related to CO₂ emissions in crop fields. These practices directly interfere on the carbon dynamics between the land and atmosphere. In this study, we investigated the temporal variability of the column-averaged dry-air mole fraction of atmospheric CO₂ (xCO₂), solar-induced chlorophyll fluorescence (SIF), and the normalized difference vegetation index (NDVI) in areas with the main agroecosystems in southern-central Brazil as a way to understand if and how crops cycle and agricultural management could be associated with the temporal variability of NDVI, SIF and xCO₂. The study was carried out in areas corresponding to the three agroecosystems': sugarcane (Pradópolis, State of São Paulo, Brazil), cropland with soybean-corn succession (Santo Antônio do Paraíso, State of Paraná, Brazil), and grassland (Águas Claras, State of Mato Grosso do Sul, Brazil). Air temperature, precipitation, NDVI, and SIF and xCO₂ were retrieved from NASA-POWER, NASA-GIOVANNI, SATVeg-EMBRAPA, and OCO-2, respectively, during a two-year study. Trends were removed from the NDVI, SIF, and xCO₂ time series applying the regression method. A negative correlation between SIF and xCO₂ was found in sugarcane and cropland areas, but in grasslands, no correlation showed up. Higher SIF values were observed in grassland (2.24 W m-2 sr-1 μm-1), and lower xCO₂ values were observed above grains, which varied from 396.8 to 404.2 ppm. Both xCO₂ and SIF followed more a seasonal pattern in sugarcane and annual crops, but over pasture this presented an unusual pattern related to higher precipitation events. Our results indicate a potential use of SIF and xCO₂ which could help identifying potential sources and sinks of the main additional greenhouse gas over agricultural areas.

Keywords: Climate changes; Climate-smart agriculture; OCO-2; Remote sensing; SIF.