MELO, E. R.; http://lattes.cnpq.br/2241293886413937; MELO, Edjair Raimundo de.
Abstract:
This research highlights the importance of water, emphasizing knowledge of the volume in
storage, as an essential proposition, in regions where there are irregularities in the amount of
rainfall and distribution over time, where due to this fact, an efficient management of this
resource, so precious, increases water availability. Remote sensing has been gaining a lot of
space in the job market, being extremely important for the management of watersheds, as it
helps to identify and monitor possible changes in natural resources. In this sense, the present
work has the general objective of analyzing the stored water volume of the public dam in the
city of Sumé - PB, through satellite images, and data with a view to assisting in the management
of water resources. The study was carried out in the public dam of Sumé -PB, which is located
in the hydrographic basin of the Alto Rio Paraíba, region of Cariri, having a maximum storage
capacity of 44.8 million m3 with the relative area of the water surface of 8.5 million m2, for the
execution of this research, orbital images from the satellites Landsat 8/ OLI, Landsat 5/TM and
ETM and Landsat 7/TM and ETM+ were used. The system used for digital processing was
SPRING version 5.5.1. For digital image processing, the following methodologies were used:
Contrast; Principal Components (bands 5, 4 and 3 - Landsat 5 and 7; and 6, 5 and 4 - Landsat
8); IVDN; Enhancement by Decorrelation; Adjusted Multispectral Compositing; and, False
color rendering (RGB). The RGB-CONTRAST was the one that presented the highest
correlation of (0.9998), followed by the CMA-DECO with (0.9979) the CMA has a value
(0.9966), whereas the CMA-CP with (0.9918 ), shows us that this method was less effective
when compared to the first methodology, which obtained much more precise and significant
data. The methodologies analyzed throughout this research were very appropriate, as all of them
obtained R2 greater than 0.99. With this, we can conclude that the correlation between the water
volumes of references and using Digital Image Processing (PDI) methodologies, were very
significant, all obtained values above 0.99, however the methodology that coincided more
positively with the data from AESA was RGB-CONTRASTE (0.9998) Therefore, the use of
simplifying methods such as those exposed in this article is of great importance in the
management of water resources, serving as an excellent mechanism for obtaining information
on the hydrological components that act in a certain region.
