Comparison of Landsat-8 and Sentinel-2 satellite images to estimate the amount of chlorophyll-a in Zaribar Lake

Abstract
Background and purpose:
Population growth and pollution caused by the discharge of various types of sewage have caused the spread of pollution and the limitation of water resources. Surface water sources are more polluted than underground water sources. This pollution leads to the increase of nutrients and the blooming of algae and their consequences, such as the increase of chlorophyll-a, change in dissolved oxygen and ultimately the reduction of water quality. By monitoring the changes in water quality, it is possible to observe, evaluate and correct the long-term trends of water quality reduction and also predict its quality changes for the future. One of the areas where a lot of sewage is discharged is Zaribar lake in Marivan city located in Kurdistan province and it was chosen as the study area due to the high amount of pollution. Due to the fact that the traditional methods of water quality evaluation are time-consuming, risky and expensive, experts use new methods that can measure the amount of chlorophyll-a. Satellite images for water quality control due to easy access in the fastest time have created the capacity to measure the amount of chlorophyll-a with high speed and accuracy compared to traditional methods.
Materials and methods:
In this research, the NDWI index was used to separate the lake from the surrounding land environment in satellite images. The use of this index is for the purpose of not separating dryness when using spectral indices that measure the amount of chlorophyll-a, they will have a high error. To estimate the amount of chlorophyll-a, spectral indices should be used that can be used on the images of Sentinel and Landsat satellites. In this research, the amount of chlorophyll-a in Zaribar Lake was checked using Landsat-8 and Sentinel-2 images in 2019 with the help of Google Earth Engine web system. For this purpose, spectral indices (2DBA, 3DBA, NDCI and FLH-Violet) were applied on the mentioned images and the predicted amount of chlorophyll-a was compared with the actual amount on the ground to find the most suitable spectral index and satellite image for estimation. Select the amount of chlorophyll-a.
Results and discussion:
The obtained results show the high accuracy of 2DBA and NDCI indices in Sentinel-2 images compared to Landsat-8. 2BDA and NDCI indices were equal to R^2 equal to 0.799 and 0.794 respectively in Sentinel-2 images, which compared to the accuracy of these algorithms in Landsat-8 images, respectively equal to R^2 equal to 0.156 and 0.125, it has relatively higher accuracy and it is proof of the high accuracy of these two algorithms using Sentinel-2 images to measure chlorophyll a compared to Landsat-8 images. The use of Landsat-8 images due to the larger size of its cells compared to the Sentinel-2 images in small areas can face a challenge in the detection of chlorophyll-a, which is clearly visible in the results, and the algorithms applied on the Sentinel-2 images have been able to obtain acceptable results. Provide approval for Landsat-8. The results showed that the use of Landsat-8 images can increase the amount of measurement error, especially in the marginal areas of the lake. Chlorophyll-a can be effective temporally and spatially, on the surface and in depth.
Conclusion:
Based on the obtained results, it can be said that using 2BDA and NDCI indices for small areas in Sentinel-2 images gives more acceptable accuracy than Landsat-8 images. One of the most important reasons is the smaller size of sentinel-2 cells. Sentinel-2 spectral images have an accuracy of 10 meters, while Landsat-8 images have an accuracy of 30 meters. In order to more accurately check the amount of chlorophyll-a in this lake, it should be checked in time series of different seasons with the help of satellite images and ground sampling, because according to the amount of precipitation in different seasons, the water level of the lake changes and the concentration of chlorophyll-a is also dependent on it. Is; Therefore, the amount of chlorophyll-a should be investigated in the conditions of low water and high water in the lake, which unfortunately was not addressed in this research due to the lack of sampling.