Water Quality Accuracy Test from Sentinel-2 Imagery in The Pasee-Peusangan Watershed Area

doi:10.21203/rs.3.rs-3323726/v1

Download PDF

Research Article

Water Quality Accuracy Test from Sentinel-2 Imagery in The Pasee-Peusangan Watershed Area

https://doi.org/10.21203/rs.3.rs-3323726/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The use of satellite imagery for monitoring water quality will be very practical as well as efficient in terms of time and cost, and can provide information on a very wide area. the purpose of this study was to identify water quality against the TSS parameter and in the Pasee-Peusangan watershed using Sentinel-2 Imagery. Evaluation was carried out through a qualitative method of a combination of NDTI and NDWI and TSS water quality parameter values. After that, a validation test was carried out using the RMSE, RE and R^2 methods. The TSS value as a water quality parameter in the Pasee-Peusangan Watershed ranges from 0 to 43.908035 (mg/l), areas with low concentrations of TSS predominate with an area of 533.868 Ha or around 98.15%. While the medium and high categories are between 0.2% and 1% respectively, the very high TSS category is 0.1% of the study area. Test Image validation test using the RMSE, RE and \({R}^{2}\)formulas, RMSE shows the suitability and accuracy of field and image data. The water quality in the Pasee-Peusangan Watershed is in the class 1 and class 2 categories, which means it is still relatively good.

NDTI

NDWI

Sentinel-2 Imagery

TSS

Watershed

Water Quality

Water is a natural resource that is important for human survival (Chadli and Boufala, 2021). One of the problems in developed and developing countries as a result of natural processes and anthropogenic activities is the problem of water pollution (Tian et al, 2019). In terms of quality and quantity, water resources have changed in a bad direction and are unable to meet the continuously increasing human needs. According to the Director of Forest and Freshwater from the WorldWide Fund for Nature (WWF) Indonesia on World Water Day 2019, almost 82% of 550 rivers in Indonesia are polluted. Water pollution in Indonesia is generally caused by agricultural waste, domestic waste and industrial waste (Sanchez et al., 2007).

Water quality assessment increases related understanding of water systems which leads to better and more effective management of water resources (Fadel et al., 2021). Spatial monitoring systems have been widely used to characterize water quality and seasonal variations and to predict future conditions (Banda and Kumarasamy, 2020). The use of satellite imagery for observing water quality will be very practical as well as efficient in terms of time and cost, and can provide information over a very wide area (Alikas & Kratzer 2017). Remote sensing is the science of obtaining information on objects, areas or phenomena without having to check directly on the location of the observation area. Remote sensing techniques are very efficient for evaluating water quality (Bugnot et al., 2018). The use of this technology and geographic information systems can quickly detect changes and provide information on water quality data with effectively and efficiently (Muhoyi H. et al, 2022).

Remote sensing technology can identify and analyze the results of recording the spectral characteristics of water with parameters that determine water quality. One of the parameters to determine water quality is the concentration of Total Suspended Solid (TSS), which is a determinant of water quality. TSS includes a wide range of materials such as biological solids, decomposing organic matter, soil and particulate matter discharged in wastewater (Wisconsin Department of Natural Resources, 2012).

This water quality evaluation method uses quantitative and qualitative methods, Normalized Difference Turbidity Index (NDTI) (Lacaux at al., 2007) and Normalized Difference Waters Index (NDWI) (Xu, 2006). Collecting data in the field cannot represent a large area to identify variations in water quality, therefore the technology can provide information to formulate policies to maintain water quality (Meyer et al. 2019). Previously, a similar study was also conducted by Ichwana et al (2022) regarding the evaluation of Erosion and Total Suspended Solid (TSS) distribution using Landsat-8 imagery in the Krueng Pasee Watershed, which is one of the watersheds in the Pasee-Peusangan Watershed. In this study there is a novelty in the use of the imagery used, namely the Sentinel-2 Image in the Pasee-Peusangan River Region. The Pasee-Peusangan River Basin is experiencing several problems including land conversion due to the development of the economic sector and affecting the surrounding environment such as water management, sedimentation and flooding. Therefore, the aim of this study was to identify water quality in terms of Total Suspended Solid (TSS) parameters and in the Pasee-Peusangan River Basin.

Tools and materials

The tools used in this study were water and soil sample measurement equipment for TSS (Total Suspended Solid) measurements to test the accuracy of image analysis results, GPS (Global Positioning System), cameras as documentation and laptops that have ArcGIS software installed and Data image Sentinel. 2.

The materials needed are Sentinel-2 Imagery downloaded from the https://earthexplorer.usgs.gov/ and https://scihub.copernicus.eu/dhus/#/home site in the Pasee-Peusangan watershed, DEM data degenerated from Sentinel-2 Imagery originating from Demnas, shapefile (SHP) area characteristic map obtained from the Ministry of Environment and Forestry.

Research Location

This research was conducted in the Pasee-Peusangan Watershed with an area of approximately 452.33 \({Km}^{2}\)located at 5°09′12″–4°49′25″ North Latitude and 96°51′27″–97°14′ 55″ East longitude which is east of Aceh Province. The research location can be seen in Fig. 1.

Sentinel-2 Image Data Processing

Cloud-free sentinel image of the study area downloaded from USGS (https://glovis.usgs.gov/). Synchronization of surface water quality assessment with atmospheric correction of satellite image data to refine water quality variables. Atmospheric correction of Sentinel-2 imagery was carried out using the Sen2Cor module (version 2.5.5) in the Sentinel-2 Toolbox which is integrated in the Sentinel Application Platform (SNAP) to remove cirrus clouds

Evaluation Method

Evaluation of water quality using Sentinel-2 Imagery for TSS by calculating the Normalized Difference Turbidity Index (NDTI) and Normalized Difference Waters Index (NDWI) values with the following equation:

NDTI = \(\frac{Red-Green}{Red+Green}\)……………………………………………………………………………..……(1)

NDWI = \(\frac{Green-SWIR1}{Green+SWIR1}\)……………………………………………………………………….............(2)

Red, Green and SWIR1 are bands, i.e. color channels on the map.

To find out the value and distribution of areas where there is sedimentation suspension, the TOA (Top of Atmosphere) is changed to BOA (Bottom of Atmosphere), then processed directly by entering the TSS algorithm in Eq. 3:

TSS = \(\left(\text{31,42} x Band 2 and Band 4 Representational Values\right)-\text{12,719}\)…………………(3)

After being modeled with the TSS algorithm, a water quality test is carried out by knowing the value and distribution of water using the NDWI algorithm with the specified spectral band or channel, namely using the red and green channels. This water index algorithm is used to determine the distribution of water quality as a first step used to obtain the distribution of suspended sediments. Next, the process of masking the area to cover an object with another object uses the algorithm test in Eq. 4.

Masking = \(if NDWI>0 then 1 else 0\) …………………………………………………..………(4)

Estimates of TSS concentrations and their validation are calculated to assess the results of the algorithm. The results of the TSS sample in the field were tested for accuracy with the Root Mean Square Error (RMSE), Relative error (RE) and Determination Coefficient (R2) equations (Jaelani et al., 2016) with the following equations:

RMSE = \(\frac{\sqrt{{\sum }_{i}^{N}{(X esti, i-X meas, i )}^{2}}}{N}\) …………………………………………………………..…………(5)

Notes :

\(X meas,i\) = each value being measured

\(X esti\) , i = estimated value of each

\(N\) = number of sample

Table 1

Range Value RMSE
Range value RMSE	Category
< 0,009	Incredibly accurate predictions
0,009 < RMSE < 0,09	Good accurate predictions
0,09 < RMSE < 0,5	Reasonable predictions
> 0,5	Inaccurate predictions
Sources: Li et al, 2013

RE = \(\frac{1}{N} {\sum }_{i=1 }^{N}\left|\frac{X esti.,i-X meas,i}{X meas}\right|100\%\)………………………………………………………..(6)

Notes :

\(X meas,i\) = each value being measured

\(X esti\) , i = estimated value of each

\(N\) = number of sample

\({R}^{2}\) = \(\frac{1}{N} {\sum }_{i=1 }^{N} {(X esti, i-X meas, i)}^{2}\) ……………………………………………………...(7)

Notes :

\(X meas,i\) = each value being measured

\(X esti\) , i = estimated value of each

\(N\) = number of sample

Table 2

Range Value\({R}^{2}\)
Range value\({\varvec{R}}^{2}\)	Category
0,86 < \({R}^{2}\) ≤ 1	Incredible
0,75 < \({R}^{2}\) ≤ 0,86	Good
0,65 < \({R}^{2}\) ≤ 0,75	Satisfy
\({R}^{2}\) ≤ 0,65	Less satisfy
Sources: Bamweyana et al, 2021

The Sentinel-2 satellite image product level is still raw data with level 1C so it still requires geometric correction to reach MSI level-2A. Geometric correction is the placement of pixel values in such a way that the results can be seen by objects on the earth's surface that are recorded by the sensor. In this research, to improve the geolocation accuracy of geometric correction and atmospheric correction using the Sentinel Application Platform, namely SNAP (Sentinel Application Platform) developed by ESA (European Space Agency).

NDWI

NDWI is a wettability index developed to describe bodies of water from satellite imagery. The NDWI wettability index is a satellite derived index from the Near-Infrared (NIR) and Short Wave Infrared (SWIR) channels which are very closely related to the water content in vegetation. NDWI is used to achieve the goal of isolating water and non-water features (water content) in the Pasee-Peusangan WS. NDWI is presented in the form of maps and tables so that it displays information about the spatial distribution of vegetation water pressure and its temporal changes over a period of time.

This study uses one image data, namely the MSI Sentinel-2 Satellite Image on June 13, 2022 in WS Pasee-Peusangan. The results of the NDWI analysis in this study are shown in Fig. 2.

The NDWI value in this study was based on Mc Feeters, S. K. (1996) which is between − 1 to 1. The designation is 0.2–1- surface water, 0.0–0.2 – Flood, humidity, -0.3–0 .0 38 – Moderate dryness, non-aqueous surface, -1 – -0.3 – Dryness, non-aqueous surface (McFeeters, 1996).

The total area of the research area based on the overlay area is 553,749 hectares, inputted from the Pasee-Peusangan watershed digitized data (shapefile). Based on the results of the NDWI classification process carried out with ArcGis 10.1, it can be seen in Table 3.

Table 3

Classification of NDWI in Pasee-Peusangan watershed
No	Category	Range	Wide (Ha)	Percentage (%)
1	Low Wetness, Non-Aqueous Surface	-1 until − 0.3	478.089	86,34
2	Moderate wetness, Non-Waterless Surface	-0.3 to 0.000001	62.926	11,36
3	humidity, flooding	0.000001 to 0.200001	6.989	1,26
4	Water surface	0.2 to 1.000001	5.745	1,04
		Total	553.749	100

Most of the Pasee-Peusangan watershed is in the category of low and moderate wetness, meaning that this area is at risk of drought. Meanwhile, a small portion of Pasee-Peusangan watershed is in the category of flood humidity which has a risk of flooding. This is in accordance with the results of research conducted by Ramadhini and Sukojo (2017), regarding the analysis of deforestation rates in North Aceh District, which is one of the areas in the Pasee-Peusangan Watershed, 2000, 2003 and 2015. The results of his research indicate that the protected forest area of Aceh District In the north, there is illegal logging and land clearing, causing land cover conversion. the effect of this conversion is the destruction of tens of thousands of hectares of forest which has an impact on the water system in this area. After obtaining visual information on water bodies using the NDWI method, the next step is the process using the NDTI method.

NDTI

NDTI is a normalized turbidity difference index which is a method that has been adapted to assess turbidity with multi-spectral remote sensing (Bid and Siddique, 2019). NDTI estimates water turbidity and is grouped into 3 categories namely low, medium and high based on the average and standard deviation of the NDTI values obtained using Satellite Imagery with the help of ArcGIS software.

The NDTI values on the map are distinguished by color, green for low sediment turbidity levels, yellow for medium turbidity levels and red for high turbidity levels. This sediment content is indicated based on the level of difference in the soil silt content at the study site. The greater the NDTI level, the greater the concentration of silt soil material in the area. The classification of NDTI values in the Pasee-Peusangan watershed can be seen in Fig. 3.

NDTI has an important role to identify and measure sediment characteristics (Bid and Siddique, 2019). NDTI was originally intended to describe water turbidity (Mondal and Bandyopadhyay, 2014). The form of turbidity can be in the form of clay soil, silt deposits, organic materials and non-organic materials (Aziz et al, 2015). Therefore, the material that causes turbidity on land is also closely related to clay or sediment matters, in Fig. 3 it can be seen that the NDTI classification in the Pasee-Peusangan Watershed shows that several areas that have high silt content are in accordance with the characteristics of this area which are in lowland with the shape of a basin, and flooding often occurs in this area.

At several other points in the Pasee-Peusangan Watershed, the area has characteristics with hilly topography and sandy loam soil types. The results showed that there was an indication of silt content with medium and high classifications. A number of river and lake networks are identified in green, which means that the level of turbidity or sediment content is at a low level.

Most of the Pasee-Peusangan Watershed, the NDTI classification is in the low turbidity range with a percentage of 80.30% of the area of the Pasee-Peusangan Watershed. The turbidity condition in this area is low from the threat of landslides and inundation. The area with clay, silt and sediment makes the area prone to landslides and flood inundation. Meanwhile, in several other areas within the Pasee-Peusangan Watershed, most of them are vulnerable to landslide threats and some points are vulnerable to flood inundation. According to Hossain (2020), there is a relationship between residue coverage and NDTI can identify agricultural land processing practices. Higher residual values are due to unmanaged agricultural practices. The NDTI range values in the Pasee-Peusangan Watershed can be seen in Table 4.

Table 4

NDTI range values in Pasee-Peusangan watershed
No	Catagory	Range	Wide (Ha)	Percentage (%)
1	Low	-0,0212 until − 0,0379	440.050,79	80,30
2	Moderate	-0.0379 sampai 0,0697	94.750,99	17,29
3	High	0.0697 sampai 0,1952	13.227,51	2,41
	Total		548.029,30	100

The NDTI value and turbidity level are positively correlated, the turbidity level describes the concentration level in units of mg/l (Bid and Siddique, 2019). One of the factors causing changes in turbidity level characteristics is reduced forest area and agricultural practices without regard to conservation techniques (Aziz et al, 2015).

TSS

Remote sensing is capable of monitoring and assessing water bodies on land by extracting water spectral information and converting it into several water quality parameters such as total suspended solids (TSS) and turbidity (Saberioon et al., 2023). In the Pasee-Peusangan Watershed, the TSS content of water bodies can be seen in Fig. 4.

From Fig. 4 it shows that several areas with high TSS content are in North Aceh District. It can be seen that the characteristics of the area are residential areas with sloping basins that are prone to flooding, so this makes a lot of sediment content there. So that the results of the research data also provide information regarding the potential and vulnerability of the soil structure contained therein. In accordance with the results of the study by Chang et al (2021) which stated that topography and land conversion variables were associated with an increase in TSS values. TSS distribution using an algorithm can provide estimated results for turbidity for the 2015 and 2019 periods from Landsat satellite imagery (Ichwana, et al., 2022). The results of the analysis of estimated TSS content in the Pasee-Peusangan Watershed can be seen in Table 5.

Table 5

The results of the analysis of estimated TSS content in the Pasee-Peusangan Watershed
No	Concentration (mg/l)	Wide (Ha)	Percentage (%)
1	0 to 0.000537	533.868,41	98,1
2	0.00161 to 0.002147	1.249,06	0,2
3	0.002147 to 8.781607	5.268,17	1,0
4	17.563214 to 26.344821	365,81	0,1
5	26.344821 to 35.126428	2.703,97	0,5
6	35.126428 to 43.908035	481,58	0,1
		543.936,99	100

Validation

To assess the performance of the algorithm obtained at Sentinel-2, it is necessary to validate the TSS content of the image processing to confirm that the field data was taken at several points in the area. Data from field measurements are presented in Table 6 below.

Table 6

Results of TSS measurements in the field
No.	Location	Coordinate		TSS Value (mg/l)
No.	Location	X	Y	TSS Value (mg/l)
1	Tengoh Seulemak village, Matang Kuli (hulu) sub-district	97,2356917	5,01402500	32
2	Jembatan Besi Parang Sikareung village, Matang Kuli sub-district	97,2672333	5,03526667	30
3	Jembatan Besi Keude Lhoksukon, Lhoksukon sub-district	97,0774833	5,14360000	31
4	Desa Meunasah Asan, Lhoksukon sub-district	97,2096361	5,11752778	51
5	Jembatan Gantung Trieng Pantang village, Lhoksukon (hilir) sub-district	97,2946222	5,09946944	49
6	Bendungan/Intake PDAM Desa Beunyot, Juli Bireuen sub-district	96,7020806	5,11524444	22
7	Jembatan Besi Desa Kubu Peusangan Bireuen sub-district	96,8043583	5,18344167	26
8	Tiengkem Manyang village. Kuta Blang (Hilir) Bireuen sub-district	96,8281556	5,21072222	26,5

The results of image processing at each location coordinate are the same as field sampling can be seen in Table 7. The TSS value obtained between the sentinel-2 image processing results and the allgorima is not much different from the measurement results from field samples. Only a few points where there is a large difference in the measurement data obtained.

Table 7

TSS measurement results based on image processing
No.	Location	TSS value from imagery (mg/L)
1	Tengoh Seulemak village, Matang Kuli (hulu) sub-district	31,4
2	Jembatan Besi village Parang Sikareung, Matang Kulu sub-district	31,6
3	Jembatan Besi Keude Lhoksukon, Lhoksukon sub-district	35,2
4	Meunasah Asan village, Lhoksukon sub-district	46,6
5	Jembatan Gantung Gampong Trieng Pantang, Lhoksukon (hilir) sub-district	50,3
6	Bendungan/Intake PDAM Beunyot village, Juli Bireuen sub-district	22,6
7	Jembatan Besi Kubu village. Peusangan Bireuen sub-district	25,3
8	Tiengkem Manyang village, Kuta Blang (Hilir) Bireuen sub-district	26,7

The results of the validation test between field measurement data and Sentinel-2 Image data are presented in Table 8.

Table 8

Table of Calculation of RMSE, RE and R² Values
Imagery data	31,4	31,6	35,2	46,6	50,3	22,6	25,3	26,7
Field data	32	30	31	51	49	22	26	26,5
RMSE	0,21
RE (%)	9,97
\({\varvec{R}}^{2}\)	0,94

RMSE provides information on the concentration of water quality parameters. RE is related to satellite-derived distribution. \({R}^{2}\) is the relationship between in situ measurements and estimated measurements and estimates of water quality parameters from TSS concentrations from Sentinel-2 Imagery. The RMSE value of 0.21 indicates that the accuracy of image data with field data is in a reasonable predictive category. The RE value (%) is 9.97 in the acceptable category. The \({R}^{2}\)value of 0.94 means it is in the very good category.

The water quality in the river appears to vary at the observation point, this depends on anthropogenic activities that are the source of pollution. Illegal gold mining and C minerals like sand and stone excavation activities can increase water pollution (Rahmatillah et al., 2021). Apart from that, erosion that occurs due to land use change can also increase total suspended sediment (Muntazar et. Al, 2021). Spatial water quality compared with field test results can be a consideration in managing water pollution with the right strategy. The use of sentinel-2 imagery in monitoring water quality can be considered because it can provide high suitability and accuracy (Sharma et al. 2018).

The TSS value as a water quality parameter in the Pasee-Peusangan Watershed ranges from 0 to 43.908035 (mg/l), areas with low concentrations of TSS predominate with an area of 533.868 Ha or around 98.15%. While the medium and high categories are between 0.2% and 1% respectively, the very high TSS category is 0.1% of the study area. Test Image validation test using the RMSE, RE and \({R}^{2}\)formulas, RMSE shows the suitability and accuracy of field and image data. The water quality in the Pasee-Peusangan Watershed is in the class 1 and class 2 categories, which means it is still relatively good.

Ethics approval and consent to participate

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The author declares that there is no conflict of interest with any party. This research has no conflicts of interest between authors and data providers

Funding

this research was supported by the Institute for Research and Community Service from the Syiah Environmental University PNBP fund with Contract No: 428/UN11.2.1/PT.01.03/PNBP/2023

Data Availability Statement

The datasets generated for this study are available on request to the corresponding author.

Author Contributions

Ichwana Ramli for Coordinating the course of the research and checking the implementation of data processing and manuscript writing; Nasrul Arahman for planning and checking article writing and checking field sampling; Atika Izzaty for checking existing data and algorithms for TSS; Firman Hadi to responses for data collection, writing research results and processing data; Tito Syajanuar for write research reports and check existing data

Alikas, K. and Kratzer, S. (2017). Improved retrieval of Secchi depth for optically-complex waters using remote sensing data. Ecol. Indic. 77, 218–227.
Azis, A., Yusuf, H., Faisal, Z., Suradi, M. (2015). Water turbidity impact on discharge decrease of groundwater recharge in recharge reservoir. Procedia Eng. 125, 199–206. https://doi.org/10.1016/j.proeng.2015.11.029
Bamweyana, I., Musinguzi, M., dan mazzi, L. (2021). Evaluation of CHIRPS Satellite Gridded Dataset as an Alternative Rainfall Estimate for Localized Over Uganda. Atmospheric and Climate Sciences, 11(4): 797–811. http://dx.doi.org/10.4236/acs.2021.114046
Banda, T.D., Kumarasamy, M. (2020). Application of multivariate statistical analysis in the development of a surrogate water quality index (WQI) for South African watersheds. Water (Switzerland) 12. https://doi.org/10.3390/W12061584
Barrett, D. and Frazier, A. (2016). Automated method for monitoring water quality using landsat imagery. Water, 8, 257. https://doi.org/10.3390/w8060257.
Bid, S., Siddique, G. (2019). Identification of seasonal variation of water turbidity using NDTI method in Panchet Hill Dam, India. Model. Earth Syst. Environ. 5, 1179–1200. https://doi.org/10.1007/s40808-019-00609-8
Bugnot, A. B., Lyons, M. B., Scanes, P., Clark, G. F., Fyfe, S. K., Lewis, A. And Johnston, E. L. (2018). A novel framework for the use of remote sensing for monitoring catchments at continental scales. J. Environ. Manage. 217, 939–950. https://doi.org/10.1016/j.jenvman.2018.03.058
Chadli, K., dan Boufala, M. (2021). Assessment of water quality using Moroccan WQI and multivariate statistics in the Sebou watershed (Morocco), Arab. Journal Geosci, 14. https://doi.org/10.1007/s12517-020-06296-5
Chang, H., Makido, Y., Foster, E. (2021). Effects of land use change, wetland fragmentation, and best management practices on total suspended sediment concentrations in an urbanizing Oregon watershed, USA. J. Environ. Manage. 282, 111962. https://doi.org/10.1016/j.jenvman.2021.111962
Fadel, A., Kanj, M., Slim, K. (2021). Water Quality Index variations in a Mediterranean reservoir: a multivariate statistical analysis relating it to different variables over 8 years. Environ. Earth Sci. 80. https://doi.org/10.1007/s12665-020-09364-x
Hossain, M.R. (2020). Study on crop residue coverage in the state of Iowa and its importance in regional water supply. Egypt. J. Remote Sens. Sp. Sci. 23, 333–345. https://doi.org/10.1016/j.ejrs.2019.01.002
I. Ramli, A. Achmad, H. Basri, A. Izzaty. (2022) Erosion and Distribution of Total Suspended Sediment (TSS) Using Landsat-8 in Krueng Pase Watershed. Proceedings of the Second International Conference of Construction, Infrastructure, and Materials: ICCIM, Lecture Notes in Civil Engineering book series (LNCE,volume 216)
Lacaux, J.P., Tourre, Y.M., Vignolles, C., Ndione, J.A., Lafaye, M. (2007). Classification of ponds from high-spatial resolution remote sensing: Application to Rift Valley Fever epidemics in Senegal. Remote Sens. Environ. 106, 66–74. https://doi.org/10.1016/j.rse.2006.07.012
Li, M. F., Tang, X, P., Wu, W., dan Liu, H. B. (2013). General Models for Estimating Daily Global Solar Radiation for Different Solar Radiation Zones in Mainland China. Energy Conversion and Management, 70: 139–148. https://doi.org/10.1016/j.enconman.2013.03.004
McFeeters, S.K. (1996). The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. Int. J. Remote Sens. 17, 1425–1432. https://doi.org/10.1080/01431169608948714
Mondal, I., Bandyopadhyay, J. (2014). Coastal Wetland Modeling Using Geoinformatics Technology of Namkhana Island, South 24 Parganas, W.B, India. OALib 01, 1–17. https://doi.org/10.4236/oalib.1100975
Muhoyi., H. W. Gumindoga., A. Mhizha b, S. N. Misi b and N. Nondo. (2022). Water quality monitoring using remote sensing, Lower Manyame Sub-catchment, Zimbabwe, Water Practice & Technology, 17 (6): 1347.
Muntazar, Joni and I. Ramli. (2021). Erosion and sedimentation analysis due to land use changes in the Krueng Pase watershed. ICATES, Vol 922 (1), 012040, IOP Conference Series: Earth and Environmental Science
Rahmatillah., H. Meilina., and I. Ramli (2021). Water quality index and the sediment criteria due to anthropogenic activity in West Aceh District, Indonesia, ICATES, Vol 922 (1), 012040, IOP Conference Series: Earth and Environmental Science
Saberioon, M., Khosravi, V., Brom, J., Gholizadeh, A., Segl, K. (2023). Examining the sensitivity of 84 simulated EnMAP data for estimating chlorophyll-a and total suspended solids in inland waters. Ecol. Inform. 75, 102058. https://doi.org/10.1016/j.ecoinf.2023.102058
Sánchez, E., Colmenarejo, M.F., and Vincente, J. (2007) Use of the water quality index and dissolved oxygen deficit as simple indicators of watersheds pollution. Ecol Indic 7:315–328
Sharma, B., Kumar, M., Denis, D. M., and Singh, S. K. (2018). Appraisal of river water quality using open-access earth observation data set: a study of river Ganga at Allahabad (India). Sustain. Water Resour. Manag. 8, 1–12. https://doi.org/10.1007/s40899-018-0251-7
Tian, Y., Jiang, Y., Liu, Q., Dong, M., Xu, D., Liu, Y., and Xu, X. (2019). Using A Water Quality Index To Assess The Water Quality of The Upper and Middle Streams of The Luanhe River, Northern China’, Science of the Total Environment. Elsevier B.V., 667, pp. 142–151.
Wisconsin Department of Natural Resources. (2012). Total Maximum Daily Load and Watershed Management Plan for Total Phosphorus and Total Suspended Solids in the Lower Fox River Basin and Lower Green Bay.
Xu, H. (2006). Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery. Int. J. Remote Sens. 27, 3025–3033. https://doi.org/10.1080/01431160600589179

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Water Quality Accuracy Test from Sentinel-2 Imagery in The Pasee-Peusangan Watershed Area

Status:

Version 1

Abstract

Figures

Introduction

Methods

Tools and materials

Research Location

Sentinel-2 Image Data Processing

Evaluation Method

NDTI = \(\frac{Red-Green}{Red+Green}\)……………………………………………………………………………..……(1)

NDWI = \(\frac{Green-SWIR1}{Green+SWIR1}\)……………………………………………………………………….............(2)

TSS = \(\left(\text{31,42} x Band 2 and Band 4 Representational Values\right)-\text{12,719}\)…………………(3)

RMSE = \(\frac{\sqrt{{\sum }_{i}^{N}{(X esti, i-X meas, i )}^{2}}}{N}\) …………………………………………………………..…………(5)

RE = \(\frac{1}{N} {\sum }_{i=1 }^{N}\left|\frac{X esti.,i-X meas,i}{X meas}\right|100\%\)………………………………………………………..(6)

Results And Discussion

NDWI

NDTI

TSS

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1