1. Modeling of water quality at the confluence region of Diyala and Tigris rivers as a function of spectral and spatial satellite data.
- Author
-
Sabeeh, Khadeeja G., Mustafa, Mustafa T., and Ismail, Alhassan H.
- Subjects
GEOGRAPHIC information system software ,WATER quality ,WATER quality monitoring ,STATISTICAL correlation ,SPECTRAL sensitivity - Abstract
This study focused on the water quality for drinking of the confluence of the Diyala River with the Tigris River, south of Baghdad. Due to the need to monitor the water quality in this area frequently and continuously, this study was conducted to find the relationship between water quality parameters and spectral bands by analyzing the data of the Landsat-8 satellite image using Geographic Information System software. The main purpose of this research was to develop a model for chemical and physical parameters based on regression analysis. The water quality parameters used in the study were included (Ph, TDS, EC, TH, Ca, Mg, Turbidity, ALK, CL, SO
4 , HCO3 , K, and Na). Samples were collected from the 16 sites at the same time and date that the satellite visited the study region, which was on 5th November 2021. The coordinates of each station were also taken using Global Positioning System (GPS) device. Band ratios had shown their effectiveness in detecting variables when there was an inverse relationship between two spectral responses to the same physical phenomenon. Then the statistical correlation between the observed and measured values of water quality variables obtained from the laboratory test and the reflectivity of the spectral bands for each pixel was used for analysis. The SPSS program was used for finding correlation and regression equations. The positive and strong correlation between water quality parameters and spectral reflectivity were helped to create 13 models for ten stations and six stations for verification purposes. The values of the correlation coefficients for determination with the model's relation ranged between (0.5-0.993), RMSE was ranged between (0.063-4.783), and NSEC was ranged between (0.581-0.993). These models could be used to predict water quality variables at any point in the study area. [ABSTRACT FROM AUTHOR]- Published
- 2024
- Full Text
- View/download PDF