Fluxo de carbono e sua variação espaço-temporal na Paraíba, utilizando sensoriamento remoto e dados meteorológicos.

Abstract

Carbon dioxide (CO₂), along with other atmospheric gases such as methane, ozone, chlorofluorocarbons (CFCs) and nitrous oxide (N₂O), play a fundamental role as a greenhouse gas, contributing to the planet's thermal regulation. These gases are also essential for photosynthesis, the process by which vegetation absorbs CO₂ and converts it into biomass. Gross Primary Productivity (GPP) quantifies the carbon absorbed by vegetation via photosynthesis and is an essential metric for monitoring the health of ecosystems. Additionally, vegetation indices such as NDVI (Normalized Difference Vegetation Index) are fundamental tools for environmental monitoring, providing information on the phenological state of vegetation and its impacts on climate and the environment. In this context, the objective of the study was to evaluate the spatial, seasonal and annual variability of the amount of CO₂ captured by vegetation in the state of Paraíba-PB, based on GPP in the period from 2003 to 2022. For this purpose, data from NDVI (MOD13A2), GPP (MOD17A2H) from the MODIS/Terra satellite, climate data from ERA5, measured at five (05) INMET meteorological stations and precipitation from 102 AESA-PB stations/stations were used. The patterns of temporal and spatial variability of NDVI and GPP obtained by MODIS products were determined. Then, GPP was estimated by the Monteith (1972) method, using data from MODIS, ERA5 and measured at meteorological stations. The results obtained showed a significant spatial and temporal relationship between water availability and vegetation development, highlighting the vulnerability of the semiarid region of Paraíba to extreme climatic events, such as severe droughts. To evaluate the data estimated by GPPMODIS with GPPMonteith 1972, statistical techniques such as the least squares method and the Willmott index were used. The correlations obtained between the GPP MODIS and the Montheith (1972) method presented the best results for the localities of Monteiro with R = 0.71 and Patos with R = 0.81, respectively. In other words, for these localities, both methods were effective in showing the variations in the GPP. On the other hand, João Pessoa, Campina Grande and Areia presented less expressive values (worse correlations) with R = 0.12, 0.39 and 0.31, respectively, suggesting a low agreement between the methods, possibly due to the effects of urbanization, since they are close to metropolitan regions. These values show that the variations in GPP calculated by the two methods are not well correlated in these cities, highlighting that the Monteith (1972) method was more efficient in capturing local variations, while MODIS offered more spatially integrated estimates. Thus, the analysis of Gross Primary Productivity (GPP) in the state of Paraíba revealed a strong dependence of vegetation on water availability, with emphasis on the vulnerability of semiarid regions. The study reinforces the importance of remote sensing for environmental monitoring and the formulation of public policies aimed at climate resilience.