Modelagens do balanço hídrico em larga escala com uso de sensoriamento remoto e dados climáticos : aplicações em áreas de crescimento agrícola da região costeira do Nordeste do Brasil

Abstract

The SAFER algorithm (Simple Algorithm for Evapotranspiration Retrieving) was applied with satellite images together with precipitation grids, to test the monitoring of water balance (WB) components in the agricultural growth region of SEALBA, agriculture growing region limited by the states of Sergipe (SE), Alagoas (AL), and Bahia (BA), in the coast of Northeast Brazil, classifying the Atlantic Forest (AF) and Caatinga (CT) biomes inside this region. The MODIS MOD13Q1 reflectance product and weather data upscaled to the 16-days satellite timescale were used to retrieve the actual (ET) and reference (ET0) evapotranspiration along the years 2007 to 2021, analyzing the average results for this period. Expressive differences on P, ET, WB (P – ET), and the evaporative fraction (ETf = ET/ET0) are noticed along the year in the region and between the biomes. The period of the highest P is in May, when the 16-day values are above 90 mm in AF and larger than 70 mm in CT. Considering the annual scale, the mean total P in CT, with 772 mm, is 83% of that for AF (927 mm). Regarding ET, the highest rates occur between July and September, when the 16-day average exceeds 2.70 mm d-1 in the FA biome and 2.90 mm d-1 in CT. At the annual scale, the average ET totals were 760 and 601 mm yr-1 for the AF and CT biomes within SEALBA, respectively, being the rates in CT being 79% of those for AF. Accounting P and ET, the CT biome, the CT biome has a slightly rainfall more water availability than the FA biome, with an annual WB being 2% larger, because even FA having a higher P, its higher ET rates reduce WB. The highest P values together with the lowest ET rates from March to August, increasing the positive WB values (P > ET) in both biomes, confirm suitability for rainfed agriculture, as well as high possibilities for rainfall water storage. However, from July to December happen the negative WB values (P < ET), indicating irrigation needs during the critical crop stages. Inferring the root-zone moisture conditions through the daily values of ETf, it is noted that the best levels are from April to September in the whole SEALBA region, with some cases the species of the CT biome with better conditions than those for AF. The lowest ETf values from October to April in CT. The lower ETf values from October to April in CT, when comparing with AF, are directly related to lower amounts of rainfall under greater atmospheric demand. Crossing the WB and ETf values it is evident a delay regarding the moment when the rains satisfy the water consumptions with that when the root-zone moisture levels reach to optimum levels after the driest periods. Although the modelling being carried out in a specific Brazilian region, the suitability of the joint use of the MOD13Q1 product and gridded Weather data shows potential for implementation of an operational system to monitor the water balance components in any environmental conditions. The large-scale accountings of precipitation, evapotranspiration, and the root-zone moisture conditions are highlighted as of strong support for public policies regarding the management and conservation of water resources, mainly under the actual scenarios of climate and land use changes, with possibility of replication of the methods to other regions from Brazil and the World, after calibrations of the modelling regression coefficients.