Analysis of Placement Pattern and Number of Rain Stations Based on The Equation of Kagan Rodda in Ciliwung Watershed

The accuracy of hydrological data is influenced by the number of rain stations in a watershed, the density and distribution patterns as well as the accuracy in recording the data itself. The rain gauge network system must be planned in accordance with the needs of the rainfall data to be collected. The placement of a rain station in a watershed (DAS) can be found using the Kagan Rodda method equation. The Ciliwung Watershed is a watershed in an area that has been developed with a high population density with a large number of rain gauges, currently there are 7 rain stations. After looking for the analysis of the placement pattern and the number of rain stations using the Kagan Rodda method in the Ciliwung watershed, it was found that 4 stations were recommended, so that the number of stations in the Ciliwung watershed became 11 stations, and the placement pattern with the Kagan Rodda triangle, the side length of the triangle is the same as the distance between rain stations. L = 7,947km.


Introduction
Accurate hydrological data will increase the accuracy of the analysis results. The accuracy of hydrological data is influenced by the number of rain stations in a watershed, the density and distribution patterns as well as the accuracy in recording the data itself. Hydrological data is a collection of information or facts regarding hydrological phenomena such as the magnitude of: rainfall, temperature, evaporation, duration of sun exposure, flow velocity, river sediment concentration will always change with time (Soewarno, 1995).
The rain gauge network system must be planned according to the need for the utilization of the rainfall data to be collected. In an area that has been developed (intensive development) with a high population density, the number of rain gauges needed should also be greater. This is because the level of development that is taking place in that place demands more accurate information about rainfall compared to less or less developed areas with low population density (Asdak, 2018).
The Ciliwung Watershed is a watershed with a large category and includes several cities and includes the capital city of Indonesia. The placement pattern and the number of stations in this watershed affect the hydrological analysis needed for the surrounding area, especially the D.K.I Province area of Jakarta, which is the capital city of Indonesia and has many government offices and business areas, and often experiences flooding.

Thiessen Method
This method is obtained by making a polygon that cuts perpendicularly in the middle of the connecting line of two rain captive posts. With each captive post the Pn rain will be located in a closed polygon area with an area of An.The average rainfall is obtained by adding up all the products of rainfall at the Pn rain breeder post with a closed polygon area with an area An for all areas located in the catchment area and then divided by the total area At, with the following equation: P = A 1. P 1 + A 2. P 2 +...+A n. P n A t Where : P = average rainfall (mm) P1-Pn = Rainfall at each station (mm) A1-An = Area delimited by polygon line (km2) At = total catchment area VOLUME 5 │ NUMBER 2 │ NOVEMBER 2020 http://adri.journal.or.id/index.php/ijens/index ISSN: 2656-1174 (online) Attribution 4.0 International (CC BY 4.0) You are free to: Sharecopy and redistribute the material in any medium or format, Adaptremix, transform, and build upon the material for any purpose, even commercially ADRI INTERNATIONAL JOURNAL OF ENGINEERING AND NATURAL SCIENCE 28

Kagan Rodda Method
Determination of the rain station network is not only limited to determining the number of stations required in a watershed, but also the location and distribution pattern. Qualitative guidance is provided by (Rodda, 1972), namely by utilizing the rainfall correlation coefficient (Sri Harto Br, 1993). This must still be related to the surrounding conditions concerning the availability of observers and their distribution patterns.
In the research conducted by (Rodda, 1972), for tropical areas where local rainfall with a very limited spread area has a variety of spaces for rain with a certain period, it is very uncertain even though it actually shows a relationship to some degree (Sri Harto Br, 1993).

Figure 1. Methodology
Based on the research flowchart, this research methodology is as follows: 1. Collecting and preparing rain data and rain stations.
a. Rain Data: Maximum daily and monthly rainfall. Rainfall data for 10 years b. Rain Station Data: area of influence of rain stations, distance between rain stations, and the number of rain stations. 2. Calculating the maximum monthly average rainfall in the watershed using the Thiessen Polygon method. 3. Calculate the distance between rain stations. 4. Calculating the coefficient of variation (Cv) from the calculation of the regional average rainfall. 5. Calculate the correlation between rainfall stations, for both daily and monthly rainfall, as needed. 6. The relationship obtained is depicted in an exponential curve graph, from this graph it can be obtained the magnitude of d (0) using the mean values of d and r (d). 7. With this quantity, the smoothing error and interpolation error can be calculated using the equations Z1 and Z3, after the high accuracy is determined. 8. After the number of stations has been determined for the said watershed, the determination of rain stations can be carried out using the equation r (d) and depicting an equilateral triangle net. VOLUME 5 │ NUMBER 2 │ NOVEMBER 2020 http://adri.journal.or.id/index.php/ijens/index ISSN: 2656-1174 (online) Attribution 4.0 International (CC BY 4.0) You are free to: Sharecopy and redistribute the material in any medium or format, Adaptremix, transform, and build upon the material for any purpose, even commercially

Rainfall Analysis
The data used to calculate the rainfall analysis comes from BBWS Ciliwung Cisadane and the BMKG Online Data Website, in the following table there is data that has been processed into maximum daily rainfall data which is then used to calculate the regional average rainfall analysis. To calculate the regional average rainfall using the Thiessen Method, the area of each rain station is required. The data on the area of the rain station are as follows : The following are the results of the analysis of the regional mean rainfall sought by the Thiessen method :

Correlation Coefficient and Correlation Radius
From the available rain station locations, you can search for the distance between rain stations and from the rainfall data for each rain station, the correlation between rain stations can be found with a regression graph. VOLUME 5 │ NUMBER 2 │ NOVEMBER 2020 http://adri.journal.or.id/index.php/ijens/index ISSN: 2656-1174 (online) Attribution 4.0 International (CC BY 4.0) You are free to: Sharecopy and redistribute the material in any medium or format, Adaptremix, transform, and build upon the material for any purpose, even commercially The correlation between the rain stations is then graphed with the exponential distance between the rain stations in order to obtain the Correlation Coefficient and Correlation Radius values.
The following is a table of distances between rain stations in Km: The following is an example of a regression graph between two stations. The following graph shows the 10-year average rainfall between each station:

Figure 2. Correlation Graph Between Stations
From the graph, the correlation value can be seen. In the example above, the correlation value is R3 of 0.3613. The following is a table recapitulation of the correlation of each station that has been searched: VOLUME 5 │ NUMBER 2 │ NOVEMBER 2020 http://adri.journal.or.id/index.php/ijens/index ISSN: 2656-1174 (online) Attribution 4.0 International (CC BY 4.0) You are free to: Sharecopy and redistribute the material in any medium or format, Adaptremix, transform, and build upon the material for any purpose, even commercially Based on the data on the distance between rain stations and the correlation between rain stations, the exponential graph can be searched to get the correlation coefficient (r (0)) and correlation radius (d (0)). The following is a graph of the exponential:

Figure 3. Correlation Coefficient Graph
Based on the graph of the relationship between station distance and the correlation between the above, it can be obtained the following equation y = 0.1814e0.014x. Where the above graph aims to obtain the Kagan parameter value, namely 0.1814 for the correlation coefficient (r (0)) and the correlation radius (d (0)) of 1 / 0.014 = 71.429.

Rainfall Analysis
To get the coefficient of variation (Cv) which will be used as the Kagan Rodda parameter, it can be obtained from calculating the mean of all rain data then divided by the standard deviation. The following is a table for calculating the coefficient of variation (Cv).
The coefficient of variation is obtained by the formula: You are free to: Sharecopy and redistribute the material in any medium or format, Adaptremix, transform, and build upon the material for any purpose, even commercially From all the Kagan Rodda parameter results that have been obtained, analysis of the rain post network in the Ciliwung watershed can be analyzed. The analysis carried out includes interpolation error, flattening error and distance between posts as well as the ideal number of posts available based on the level of error. The following is the calculation formula and table recapitulation: Based on the table above, the number of rain stations in the Ciliwung watershed is 7 stations with an average error value of Z1> 5%. Furthermore, it can be drawn an equilateral triangle with sides equal to the distance between rain stations (L). The following is a calculation of the distance between rain stations (L). L = 1.07 √ A n = √ 386.103 7 = 7.947 km From the calculation of the length of the distance between rain stations, the value of L = 7.947 km is obtained. Next, a triangle is drawn with L = 7,947 km on the Kagan Rodda network map to determine the distribution of rain stations in the Ciliwung watershed. You can see the following image of the distribution of the rain station: