Assessment of groundwater pollution risk potential using DRASTIC Model in Yogyakarta City, Indonesia

Assessment of groundwater pollution risk is purposed to be of particular value for environmental management programs development, this research estimated and calculated of groundwater pollution risk potential in Yogyakarta City using DRASTIC Model, DRASTIC parameters developed as Depth to water (D), net Recharge (R), Aquifer media (A), Soil media (S), Topography (T), Impact of vadose zone (I), and Hydraulic conductivity (C), the weights and ratings assigned in this research were similar to the typical ratings suggested in the original Methode. The result final assessment that groundwater pollution risk potential are classifies 4,27% of the area as having very low, 15,12% of the area as having low, 20,17 % as having moderate, 43,10% as having high, and 17,34% as having very high. The model was validated using E. Coli parameters as a represent of pollution risk where the results having exceeds 85% of total samples that a value exceeds from the standard of quality > 0 MPN/100 ml. However, this research is a preliminary to be continued in the further research.


Introduction
Yogyakarta city is an urban environment area in Indonesia, the population growth will increase from 422,732 people to 500,000 in 2030 with an average population growth of 1.04% [1]. In addition, the pressure of population factors from tourism activities where a number of tourist visits recorded in 2017 there were 3.89 million tourists with a proportion of foreign tourists (11.2%) and domestic tourists (88.8%) and an average length of stay is two days. Based on population and number of tourists, that can be implied that the number of people reached 1 million for the average mobility of people for a month. This number given the significant development of urban developments in the city of Yogyakarta, it is necessary to evaluate comprehensive the changes in the urban environment in Yogyakarta City and ensure the sustainability, especially the quality of groundwater. In principle, the sustainability of an urban area can be seen from the adaptation process in maintaining a balance between ecosystem services and human well-being inside area. Sustainability includes three components of the environment, namely abiotic, biotic and cultural components. Environmental quality has a fundamental role in influencing the sustainability and sustainability of environmental quality degradation and vulnerability [2]. Therefore, in order to see the extent to which changes or dynamics of environmental quality can be evaluated through various methods. Dynamics of environmental quality can be evaluated especially the quality of groundwater by DRASTIC method, as like similar index 2 Materials and Methods

Research area description
The research area is located in Yogyakarta City, show in Figure 1. The research area has complex characteristics with total area ± 32.53 km 2 . Population in Yogyakarta City is one of the most densely populated around 13,007 in habitants/km 2 . The city consists of 14 districts and has a border with Sleman and Bantul regency. The development of the Yogyakarta city is generally influenced by the education, services and tourism.

The DRASTIC model
The DRASTIC model is used to evaluate groundwater vulnerability index to indicate pollution risk potential. This method describes the seven (7) parameters that are analysed with their abbreviations as defined according to previous research [3,4,5] shown in Table 1.
Where DI, is vulnerability index (D, R, A, S, T, I, and C are the seven parameters) as defined in Table 1; and the subscripts r, i and w are the parameters rating for grid cell i and weights.
Data came in various spatial resolutions shown in Table 2. The weights have relationship the relative importance of each parameter with respect to the other parameters. The weights parameter is constant for each parameter the designated rating varies from 1 to 10 and the rating ranges were determined depending on the properties at Yogyakarta City scale. the weights and rating ranges of the parameters shown in Table 3 dan  Table 4. In general, the weights and ratings assigned in this research were similar to the typical ratings suggested in the original DRASTIC method.

Data base compilation
All data parameter (hydrogeology, geology, soil, groundwater recharge and topography) constructed by GIS database in Yogyakarta City scale. The metadata of the constructed GIS database, Data came in various spatial resolutions, therefore data layers to be suitable with the proposed resolution of the GIS model based 1: 35.000 scale for this research

Development of DRASTIC parameters
An overview of DRASTIC parameters develop where is each parameter as Depth to water (D), net Recharge (R), Aquifer media (A), Soil media (S), Topography (T), Impact of vadose zone (I), and Hydraulic conductivity (C) processed in the GIS is described below.

Depth of groundwater (D)
Term Depth to water table for initial D, This paramater refers to determinate of the vertical distance from the land surface to vadose zone in the aquifer. It can be described the distance that a potential contaminant migration to travel before end up to the aquifer. In general, the depth to groundwater is properly variable with respect to groundwater sensitivity, because the distance a contaminant flow through to the unsaturated zone before reaching the water table [7]. The distribution of the depth of groundwater parameter was inferred by subtracting the groundwater level from 80 (eighty) wells over the research area as presented by Public Works-Energy and Mineral Resource Office of D.I Yogyakarta. The weights and ratings for I are shown in Table 3.

Net Recharge (R)
Term net recharge for initial R. This paramater refers to represents the amount of water per unit area of the land surface to vadose zone in the aquifer. The net recharge is influenced by the amount of surface runoff, the slope, and the permeability of the soil, the amount of water and the recharge of aquifer [8]. The net recharge parameter was derived by BMKG D.I Yogyakarta. The weights and ratings for I are shown in Table 3.

Aquifer Media (A)
Term Aquifer media for initial A, this paramater refers to determinate type of consolidated or unconsolidated material in the aquifer. In this research area, the lithological represents aquifer media were determined by Yogyakarta formation, which is a result of Merapi Volcano ejecta in Quaternary succession and composes of sand, gravel, silt and clay and Sentolo Formation, occurred in Tertiary succession and consists of marl, limestone and marly limestone [9]. The categories are unconsolidated sediments, according to the regional geological context, the research area are located in middle part of Yogyakarta basin. The vulnerability of the aquifer will increase if the grain size and the fractures or openings within the aquifer [10]. The Aquifer media were inferred by Public Works-Energy and Mineral Resource Office of D.I Yogyakarta. he weights and ratings for I are shown in Table 4.

Soil Type (S)
Term soil type for initial S, This paramater refers to represents that media the contaminant passes through when it percolates into vadose zone in the aquifer. According to impact soil type for contaminant [11], Soil type has a significant impact on the amount of recharge that can be infiltration into vadose zone in the aquifer and hence on the ability of a contaminant to move vertically. The soil type of Yogyakarta City was inferred from the data processed by Agency For Agricultural Research and  Development where the uppermost part of study area is covered mostly by sand and sandy loam and greater grain sizes of geologic materials such as sand, coarse sand, and gravel, The weights and ratings for I are shown in Table  3.

Topography (T)
Term Topography for initial T, this paramater refers to defines the runoff and infiltration capacity of the land surface into the soil through into vadose zone in the aquifer. The slope is an important parameter, which is more runoff will be generated and hence groundwater contamination risk will be low [3]. Therefore, it will be increasing the potential for introduce of contaminants. The slope were inferred by Public Works-Energy and Mineral Resource Office of D.I Yogyakarta and resampled and reclassified with the ratings into six classes in the research area. The weights and ratings for I are shown in Table 3.

Impact of vadose zone (I)
Term impact of vadose zone for initial I, this paramater refers to influences the residence time of pollutants in the unsaturated zone [7]. The role of the unsatured zone above the water table is integrated with respect to vadose zone. The vadose zone were inferred by Public Works-Energy and Mineral Resource Office of D.I Yogyakarta It The weights and ratings for I are shown in Table 4.

Hydraulic conductivity (C)
Term the Hydraulic conductivity for initial C, this paramater refers to measure of the ability aquifer for transmitting water when submitted to a hydraulic gradient that described how pollutants transfers by travel of velocity the pollutants, and hence the residence time and dilution which is a high conductivity will be related to high contamination pollution risks [10]. Aquifer media refers to the consolidated or unconsolidated material in the aquifer. then, hydraulic conductivity refers to the ability of the aquifer materials to transmit water, which is the rate can be controlled that groundwater will flow under a given hydraulic gradient [9]. Therefore, hydraulic conductivity is important parameter because it can be described for dilution which a contaminant will move flowing through into vadose zone in the aquifer. The hydraulic conductivity was inferred by Public Works-Energy and Mineral Resource Office of D.I Yogyakarta.
The weights and ratings for C are shown in Table 3.

Development of groundwater pollution risk
The groundwater pollution risk related to the potential of a groundwater body for undergoing groundwater contamination [12]. The risk of pollution is determined both by the groundwater vulnerability index of the aquifer, which is relatively static, and the existence of potentially polluting activities at the soil surface. These latter activities are time dynamic and can be controlled [9]. Population density is often used as a proxy for pollution pressure at the soil surface and groundwater plays crucial role in urban environment as around 70 % of population are using shallow wells. However, groundwater quality has been tremendously threatened by using on-site sanitation systems . In this research, the population density was inferred by Civil Registry Service Office of D.I Yogyakarta. The population density is grouped into five classes into very low, low, moderate, high and very high by generated the groundwater pollution risk index by combining the groundwater  (2) where MDI is the modified DRASTIC index for risk assessment, Di is the DRASTIC index and Pw × Pr is the multiplication of rating for grid cell I and weight for population density, using the following the equation: Di, max -Ui,min where Pr is the multiplication of rating for grid cell I for population density and Pw is the weighting for population density show in Table 5.

Results and discussions
The result assessed the groundwater vulnerability index and groundwater pollution risk potential in Yogyakarta City scale, the result deployed the empirical of DRASTIC method into a GIS by is processed and overlay with ArcGIS 10.2™ and sensitivity parameter analysis is presented by Geoda TM .

Mapping of groundwater vulnerability index
The groundwater vulnerability index was calculated (rating × weight of each parameter) by the addition of the different parameters, the resultant calculated each parameter. The groundwater vulnerability index map is show in Figure 2. Which is classified by 5 classes, with higher values describing greater vulnerable. The groundwater vulnerability index indicated that southern and a portion of the middle are dominated by having very high and high vulnerable. It is noticed that unsaturated zone in middle part having high vulnerable while most of southern having very high vulnerable, there is located in Mantrijeron, Mergangsan, Pakualam and Gondomanan district. This might be due to the reason that this area is mainly covered by coarse sand and mostly the settlement/tourism activity. The resultant classes index is shown in Table 6 where classes have been grouped overall into vulnerability interval index.

Statistical Summary of DRASTIC parameters
A statistical summary of the seven (7) parameters is shown in Table. 7. This statistical summary presented the significance of each parameters in the assessment of the groundwater vulnerability index. On average rate values, the T parameter (mean=9.38) has the highest number, the D (mean = 9.32), R (mean = 9.00), and A (mean = 8.83) parameter have a high number. The I (mean = 8.00) and S (mean = 7.51) parameter have a moderate number. The C (mean = 3.67) parameter imply a low number. The coefficients of variation (CV) clarified that a high contribution to the variation of groundwater vulnerability index is purposed by R (92.30%), S (76.49%) and D (65.16%). Having moderate contribution is clarified with respect to variability of C (42.67%) and A (39.62%), and having low contribution is clarified with respect to variability of I (20.05%) and T (18.53%). In this research area, vadose zone and aquifer media are composed of the same material, A and I parameter have the same minimum values (Min = 8) as value S and A the same minimun value (Min= 2) expected to be the lowest.

Parameter sensitivity analysis
The single parameter sensitivity analysis allows the comparison between the effective and theoretical weights is shown in Table. 8. The effective weight of the DRASTIC parameter is a function of the theoretical weight and the interaction with the other parameters. The effective weight of the DRASTIC parameters obtained in this research exhibited some deviation from the theoretical weights. The Depth of groundwater (D) parameter tends to be the most effective parameter where D parameter has the highest effective weight values of the groundwater vulnerability index (16,73%). The effective weight of the R, I, C parameter (16.15%, 14,36%, 6,58%)

Mapping of groundwater pollution risk
The result of groundwater pollution risk potensial assessment is shown in Figure 3 and Table 9. The map classified into five classes to a very low, low, moderate, high and very high groundwater pollution risk potential. The result assessment that the final groundwater pollution risk potential for the research area classifies 4,27% of the area as having very low pollution risk potential, 15,12% of the area as having low pollution risk potential, 20,17 % as having moderate pollution risk potential, 43,10% as having high pollution risk potential, and 17,34% as having very high pollution risk potential. We identified high to very high pollution risk classes, area having a low pollution risk does not having pollution risk, at least it is not significant potentially from groundwater contamination, but it is relatively less susceptible to contamination compared to other district. The density population in research area is expected due for explaining the risk potential. The groundwater pollution risk index indicated that southern and middle of area research are dominated by very high and high risk pollution potential (Danurejan, Gondomanan, Mantrijeron, Mergangsan, Pakualam and Gondomanan District). Therefore, assesement of groundwater pollution risk can be explained the the evidence of pollution (biological and/or chemical) and associated health problems were identified through water quality analysis [9] [12].

Validation of groundwater pollution risk
The above mentioned the DRASTIC method is an indirect method for evaluating groundwater pollution risk potential. Indeed, the use of methods that are not validated can result in erroneous conclusions. We measured quality of water, a total of 100 samples were taken from the water well. We selected parameters E. Coli concentrations in groundwater as a proxy since anthropogenic activities by urban development in Yogyakarta City show in Table.10  and Table.11. E. Coli as an indicator of groundwater pollution whereas E. Coli was divided into two categories based on guidelines for drinking water are 0 MPN/ml and > 0 MPN/100 ml [13] [14]. E. Coli concentrations came from feces to prove the increase of E. Coli concentration in the settlement since anthropogenic activities by urban development in Yogyakarta City [15]. Based on collected data E. Coli concentration as an indicator of groundwater pollution, it can be concluded that groundwater in research area has been polluted by E. Coli parameter through 85% of total samples has exceeded the quality standard for drinking water. According to the regional groundwater flow on the research area, groundwater flow due from north to south with the slope of water table that gradually diminishes [16]. Based on evaluating groundwater pollution risk potential shown in Figure 3, it can be concluded that inferred the consistency between the calculated groundwater vulnerability and groundwater pollution risk as using E. Coli concentration data as proxies.

Conclusion
The results research assessed the groundwater vulnerability for pollution risk in Yogyakarta City scale, our research identified the seven of each parameters DRASTIC Method by Depth to groundwater (D), Net Recharge (R), Aquifer media (A), Soil media (S), Topography (T), Impact of vadose zone (I), and hydraulic Conductivity (C) from available data, and compiled the datasets into 1: 35.000 scale. We overlay the groundwater vulnerability index with population density parameter to Fig. 3. Groundwater pollution risk potential map assess the groundwater pollution risk potential. The result final assessment that groundwater pollution risk potential are classifies 4,27% of the area as having very low ; 15,12% of the area as having low, 20,17 % as having moderate, 43,10% as having high, and 17,34% as having very high pollution risk potential. The groundwater pollution risk potensial was validated using E. Coli concentration. High value E. Coli concentrations detected in high vulnerable and high pollution risk classes. The model was validated using E. Coli parameters as a represent of pollution risk where the results having exceeds 85% of total samples that a value exceeds from the standard of quality > 0 MPN/100 ml. This data inferred the consistency between the calculated vulnerability and groundwater pollution risk.