Identification of groundwater potential zones in North Province of Sri Lanka using GIS and Remote Sensing.

Introduction

Water is fundamental for all life forms on Earth, supporting both the economy and social development. Groundwater, which is found beneath the Earth’s surface, is one of the most reliable sources of fresh water, especially in regions where surface water is scarce. It plays a critical role in agriculture and domestic water supply, making it essential to understand where it can be found and how it can be sustainably managed.

Groundwater mapping helps to identify areas where water can be found and is vital for ensuring its effective management. Technologies like Geographic Information Systems (GIS) and remote sensing are used to analyze geological and environmental data to create detailed maps showing regions with the best groundwater potential. For example, research by Waikar and Nilawar (2014) demonstrates how GIS tools can integrate data on land slope, geology, and drainage density to map areas where groundwater is likely to be stored. These techniques make it easier for policymakers to pinpoint where to extract groundwater and how to protect it from overuse.

In another study, Lee et al. (2020) used machine learning alongside GIS to predict groundwater potential in South Korea. By using factors like soil type, land cover, and topography, their model was able to create highly accurate maps of groundwater availability. This advanced approach allows regions with limited data to still make informed decisions about water management

Problem Statement

Limited surface water availability and higher demands for irrigation and domestic use us water, is caused for the challenge with access to portable water in North Province of Sri Lanka. Due to this groundwater has become a critical resource in this region. And the distribution of groundwater in the Northern Province is uneven, creating challenges in identifying suitable locations for groundwater extraction. The people and animals live in regions with same conditions to North Province of Sri Lanka are affected, mainly during drought periods.

An existing spatial information or detailed mapping to identify groundwater potential zones in the Northern Province of Sri Lanka is unavailable. Inefficient groundwater extraction, overexploitation and incorrect boreholes placement in areas with low groundwater yield are occurred due to insufficient data.

In here focus to use GIS and remote sensing technologies to map groundwater potential zones in the Northern Province of Sri Lanka to overcome the information gap, that enable to make better decisions for groundwater extraction and management.

Justification

In semi-arid regions like North Province of Sri Lanka the groundwater is considered as a critical resource, where the unevenly distribution and limited availability of surface water. To ensure the sustainable water usage and management for agriculture, domestic use, and industrial purposes, need to identify the potential zones of ground water. The mapping of groundwater resources in more effectively, is facilitated by the improvement of Geographic Information Systems (GIS) and remote sensing technologies.

The effectiveness of the usage of groundwater resources is depended on sustainable management of resource. To obtain this sustainable management, policymakers, authorities, researchers, and other stakeholders must have access to accurate and reliable information. These type ground water potential maps give required information, that helps in sustainable groundwater management and consumption

Research materials and methods

Access data and sources

This study was conducted in the Northern Province of Sri Lanka, using various data sources, including Digital Elevation Model (DEM) maps and landsat satellite imagery, where obtained from the United States Geological Survey (USGS) Earth Explorer website (https://earthexplorer.usgs.gov/), and soil map of world is used

Soil map data came from the Food and Agriculture Organization (FAO) and can be found on their FAO Soil Portal. https://www.fao.org/soils-portal/data-hub/soil-maps-and-databases/faounesco-soil-map-of-the-world/en/

Digital elevation model was downloaded for drainage and slope analysis, which will help derive drainage density and slope of study area. (https://earthexplorer.usgs.gov/)

This landsat images be used to prepare land use data. (https://earthexplorer.usgs.gov/

Rainfall data in grid format was obtained from the Climate Research Unit (https://crudata.uea.ac.uk/cru/data/hrg/).

Datasets used

Downloaded data

Data type	Resource
Digital Elevation Model (DEM)	United States Geological Survey available at https://earthexplorer.usgs.gov/
Landsat satellite images	United States Geological Survey available at https://earthexplorer.usgs.gov/
Rainfall data	Climate Research Unit available at https://crudata.uea.ac.uk/cru/data/hrg/
Soil map of the world	FAO website at FAO/UNESCO Soil Map of the World \| FAO SOILS PORTAL\| Food and Agriculture Organization of the United Nations https://www.fao.org/soils-portal/data-hub/soil-maps-and-databases/faounesco-soil-map-of-the-world/en/

Processed data

Data type	Source
Elevation map	Processed from Digital elevation model
Drainage density map	Processed from Digital elevation model
Slope map	Processed from Digital elevation model
Rainfall map	Processed from Rainfall data
Land use and land cover map	Processed from Landsat satellite images
Soil map	Processed from world soil map

Data processing

WGS_1984 coordinate system was used for the entire study.

The Digital Elevation model (DEM) were converted to a mosaic utilizing ArcGIS through the data management tools (mosaic to raster) in order to come up with a single DEM. Using hydrological tools to obtain drainage density and surface tool for slope analysis.

Land use and land cover data combined and make land use and land cover map of North Province of Sri Lanka. Soil map was created using clipping world soil map into study area. Rainfall map was created using interpolation techniques

Elevation Map

Digital Elevation Model which was downloaded from USGS Earth Explore, was used to extract the elevation map of North Province of Sri Lanka. Downloaded parts of DEMs were integrated using Mosaic to new Raster tool in Data Management tools to obtained the complete DEM of North Province of Sri Lanka.
To clip the complete DEM into study area, use the Clip raster tool in raster processing of Data Management tools
Obtained the Elevation map of North Province of Sri Lanka

Drainage Density Map

Complete DEM was analyzed using hydrology tool in Spatial Analysis tools of ArcGIS toolbox. Fill tool was used to limit the amount of probable inaccuracies of DEM, which adjusted the elevation.
Flow direction was calculated using the Flow Direction tool in Spatial Analyst Tools Hydrology Set and it took input of the filled DEM.
The Flow Accumulation was prepared using Flow Accumulation tool in same set of hydrological tools and input raster as the flow direction raster prepared previously.
In a digital elevation model (DEM), the Flow Accumulation raster shows how many cells are upstream from a specific cell, indicating how much water might flow through that cell. Cells with Flow Accumulation values greater than 500 are considered to have a high potential for water flow. In the new raster created, we assign a value of 1 to these cells (those with values over 500) and a value of 0 to all other cells. This process helps identify potential drainage lines or channels that are likely to carry significant amounts of water. This analysis was performed using the CON tool in conditional tools. Use Dem file as true raster file.
The Stream Order tool in Hydrological tools of ArcGIS helps categorize streams in a watershed based on their position in the overall drainage system. It uses the Strahler method, where the smallest streams are given an order of 1. When two of these first-order streams merge, they form a second-order stream, and this process continues as streams combine. The tool produces a map (raster layer) that assigns a specific order to each stream segment, which helps in understanding the structure and flow patterns of the drainage network. Use conditioned flow accumulation file as input.
The stream to Feature Tool Hydrological tools of Spatial Analysis tools takes stream data stored in a raster format and converts it into vector line features. The result is a more flexible format, like a shapefile or feature class, that represents the streams as lines. This makes it much easier to view, work with, and analyze the stream data in various ways. Stream order file used as input.
The Line Density tool in Density tools calculates how concentrated streams, are in a given area. It looks at a neighborhood around each cell in a grid and measures the total length of lines within a set distance. The result is a map that shows areas with more or fewer of these features, making it useful in hydrology to visualize the density of stream networks across a study area. Need to set processing extent and mask as the study area.
Obtained the Drainage Density map of North Province of Sri Lanka.

Slope Map

The clipped complete DEM file was used to make slope map of North Province of Sri Lanka.
Slope tool in Surface tools of Spatial Analysis tools was used to obtain the slope map.

Rainfall Map

Multidiamentional raster data as input
Rainfall data was downloaded from the Climate Research Unit (CRU) that provided in a grid format. Since the data was converted into a more usable format. Each band in the dataset represented a month, so the data for each month was separated, and then summed to create an annual rainfall map. The Cell Statistics tool in Spatial Analysis tools was used to calculate the total rainfall for the year by adding up the last 12 bands, each representing one month’s rainfall.
The next step involved converting the rainfall data from a raster format to point data, with each point representing the total rainfall at that specific location, to obtain this the raster to point tool was used. IDW Interpolation tool in Spatial Analysis tools was used for estimating rainfall across the area, as it handles complex and changing rainfall patterns effectively.
Clip Raster tool was use to obtain the rainfall map of North Province of Sri Lanka.

Soil Map

The world soil map shape file was used to make soil map of North Province of Sri Lanka.
The world soil map was clipped into study area shape file (North Province of Sri Lanka), and the soil map of North Province of Sri Lanka was obtained.

Land Use and Land Cover Map

Downloaded Landsat images were used to obtain the land use and land cover map of North Province of Sri Lanka.
Composite bands tool in Data Management tools was used to make same number of bands in each downloaded satellite image. The Mosaic New Raster tool in Data Management tools was used to obtain the complete salute image of North Province of Sri Lanka. Need to input number of bands in the satellite images to complete process. The obtained satellite image was clipped into study area using Clip raster tool.
The Training Sample Manager in Imagery Classification Tools was usedto train the satellite image with each land use and land cover types. And saved the trained sample.
Classify tool in Imagery classification tools used with Maximum Likelihood Classification and the trained segmented file was used as input file.
Obtained the Land Use and Land Cover map of North Province of Sri Lanka.

Data analysis

Data reclassification

All prepared maps were reclassified separately for in order to prepare for overlay analysis. Use reclassify tool in Spatial Analysis tools. Groundwater potential zones were needed which led to the creation of 5 classes.

Weighted overlay analysis

The overlay analysis tool uses two key ideas which were influence and scale value. “Influence” refers to how important each layer is, and the total influence should always add up to 100%. For example, the land use layer might have a certain influence. The “scale” refers to how important specific land use type within a layer are, with ratings like 1 being less important and 5 being more important. To do the analysis, Weighted overlay tool in Spatial Analysis tools was used. All the prepared raster maps such as reclassified drainage density, slope, digital elevation model, rainfall, land use, and soil map were set the evaluation scale from 1 to 5 to create five classes for groundwater potential zones. Areas with higher weights are considered to have more groundwater potential.

Weighted overlay influence assignment based on importance of layers

Raster map	Influence (%)
Elevation map	30
Drainage density map	30
Slope map	5
Rainfall map	15
Land use and land cover map	10
Soil map	10

Drainage Density (30%): Higher number of streams occur in an area, make more water infiltration and generate higher ground water recharge. The scale values were set 5 for areas with high drainage density and 1 for low drainage density.

Slope (5%): Flatter areas are better The groundwater recharge was higher in low sloppy areas and the scale value 5 was assigned or flat areas and 1 was assigned for higher sloppy areas, because steeper slopes cause more water to run off quickly, leaving less opportunity for it to soak into the ground.

Elevation (DEM) (30%): Lower elevations can collect more water, which may contribute to groundwater, 5 was assigned or lower elevations and 1 was assigned for higher elevations

Rainfall (15%): Higher rainfall areas make higher infiltration an ground water recharge, these areas assigned as scale value 5 and the lowest rainfall areas assigned as value 1

Land Use (10%): Different land uses, such as water bodies, agricultural plantations, bare lands and urban developed areas, divergently affect the water infiltration. The infiltration was ascendingly ordered as water bodies, agricultural plantations, bare lands and urban developed areas, the scale values

2 – Urban developed areas (lowest infiltration)
3 – Bare lands
4 – Agricultural plantations
5 – Water bodies (highest infiltration) were assigned.

Soil Type (10%): Certain soils, like sandy soils, allow more water to pass through and reach the groundwater, but this factor is generally less important than rainfall or drainage.

RESULT AND DISCUSSION

The results show that high groundwater potential areas cover about 45 square kilometers. Most of these high ground water potential zones are found in the southern part of North Province of Sri Lanka, while regions with poor to moderate potential are located in the northern parts of North Province of Sri Lanka. The combination of different tools in ArcGIS was used to obtain the final map illustrating different groundwater potential levels.

Groundwater potential zones are influenced by various factors in the area, such as elevation, land use, soil, drainage density, slope and rainfall distribution. Among these, elevation and drainage density play the biggest roles. Elevation is closely linked to groundwater presence, areas at higher elevations tend to have less groundwater, while lower elevations tend to collect more ground water. For drainage density, areas with more streams allow more water to infiltrate the ground, leading to higher groundwater recharge, while areas with fewer streams experience less infiltration.

When consider slope, more steep slopes make high runoff rates, that create low water infiltration for ground water, as well low slope areas allow to maintain high potential for ground water recharge and accumulation. The parameter rainfall, moderately effects on ground water availability, areas with higher annual rainfall helps to maintain higher ground water potential, as well, areas with low annual rainfall, make low ground water potential.

Land use and land cover is another parameter that effects on ground water potential, and high water available areas, like water bodies, make high water infiltration and higher ground water potential. Followed by agricultural plantations, bare lands and urban developed areas with gradual decrease of water infiltration and ground water availability.

Soil type of the study area, affect on the ground water potential. Sandy soils allow to higher infiltration of water and make higher ground water potential as well, soils with more clay content, allow to less water infiltration and make low ground water potential.

Reclassify tool in Spatial Analysis tools was used to simply the obtained maps into five classes. Obtained classified maps were combined by using Weighted Overlay Analysis tool with setting influence percentages of each parameter map and set the 1 to 5 scale for each feature of each parameter map to performs overlay analysis and generate ground water potential map of North Province of Sri Lanka.

Areas with poor ground water potentials, in the northern part, mainly due to low streams occurring in this area, and the higher ground water potential zones in southern part, mainly due to higher availability of streams and higher amount o annual rainfall in that area. Drainage density makes the main role for the ground water potential zone identification.

Overall the resultant ground water potential of each area is based on the different factors.

CONCLUSION AND RECOMMENDATIONS

Conclusion

Identifying groundwater potential zones using GIS and remote sensing in North Province of Sri Lanka, that this approach has become a suitable tool for evaluating groundwater resources, especially in arid and semi-arid areas. The study demonstrated that combining GIS and remote sensing techniques offers an efficient way to identify and map potential groundwater zones in the area. It emphasized the importance of considering various factors like elevation, land use, soil type, slope, drainage patterns, and rainfall in the mapping process. The research also highlighted how remotely sensed data, like Landsat images, can be used to detect and classify geological structures, vegetation cover, and other surface features and other parameters that play a key role in groundwater occurrence and flow.

Utilizing GIS and remote sensing technologies can greatly improve the accuracy of identifying areas with groundwater potential, leading to more effective management and conservation strategies or ground water. This study providing valuable insights for decision-makers to promote sustainable groundwater use and develop comprehensive analyses of ground water potentials, and these tools play a crucial role in accurately mapping groundwater potential zones in study area.

This study gives important insights into how GIS and remote sensing methods can be effectively used for mapping groundwater potential. The findings could help shape groundwater management policies, improve water supply planning, and support sustainable land-use planning in areas like Northern Province of Sri Lanka and similar regions. Overall, the study’s results are valuable for environmental management, sustainable development, and resource conserve ation, especially in dry and semi-dry regions.

Samatanga, S. (2024). Identification of groundwater potential zones using GIS and Remote Sensing: A case study of Chitungwiza. ResearchGate. https://doi.org/10.13140/RG.2.2.14489.19046