Spatial and Temporal Analysis of the Water Quality in Brazilian Semi-Arid Reservoirs Using AlgaeMAp
摘要
The Northeastern Semi-Arid region of Brazil grapples with intricate water quality challenges given the prolonged dry periods, high temperatures, and limited vegetation cover. As a result, the semi-arid reservoirs are generally prone to eutrophication and algae blooms, but comprehensive insights into water quality conditions across the region remain elusive due to lack of regular information on water quality. Alternatively, the use of remote sensing has proven an effective tool on estimating key water quality parameters, such as chlorophyll-a concentration (Chl-a). In this chapter, AlgaeMAp, an application that uses NDCI (Normalized Difference Chlorophyll-a Index) derived from Sentinel-2 imagery on the Google Earth Engine platform, is applied to estimate Chl-a, and Trophic State Index (TSI). The aim of this research is to comprehensively investigate water quality conditions, specifically Chl-a and TSI (Trophic State Index), for 30 reservoirs located in 4 federal states. First, a validation of the Chl-a and TSI estimations provided by the AlgaeMAp application was conducted by comparing them with in situ measurements. Second, a spatial and temporal analysis of water quality, including Chl-a time series and its trending overtime, for the 30 reservoirs was conducted. Further, relationships between water quality and hydrological conditions are given, including precipitation and dams’ water levels, offering insights into the dynamics of trophic states for Castanhão reservoir. We have demonstrated that AlgaeMAp serves as a useful tool for analyzing Chl-a and TSI in Brazilian semi-arid reservoirs. Despite the data initially not being intended for remote sensing applications, the Chl-a estimations demonstrated acceptable error values, supported by a high determination coefficient (R2 = 0.89, n = 188). We found that the reservoirs along the São Francisco River exhibited oligotrophic conditions, but with a gradual transition toward mesotrophic conditions, which could pose concerns for future water resources within the region. Considering all reservoirs investigated, 23 out of 30 exhibit low water quality (eutrophic or higher) throughout 2017 and 2022 indicating a scenario of degraded water condition. On the other hand, examining time-series data, we observed that 15 out of 30 reservoirs displayed a decreasing trend in Chl-a concentration. Lastly, we described the influence of hydrological effects on the water quality particularly for Castanhão reservoir; it is important to note that a severe drought persisted until 2018 causing water degradation in Castanhão and most of the reservoirs. With successive rainy periods, this reservoir has recovered some water storage resulting in water quality improvements. However, it is evident that there is no uniform temporal pattern applicable to all reservoirs in the region and further research is suggested.