Open Access proceedings Journal of Physics: Conference series Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 18 Hydrological Analysis of TRMM (Tropical Rainfall Measuring Mission) Data in Lesti Sub Watershed1 Suryaningtyas Lufi1, Suhartanto Ery2, Rispiningtati2 1Hydrology Department, PT Saka Buana Yasa Selaras, Banjarmasin, 70236, Indonesia 2Water Resources Engineering Department, Universitas Brawijaya, Malang, 65145, Indonesia lufisuryaningtyas@gmail.com Received 19-12-2019; accepted 02-03-2020 Abstract. Alternative solution for availability of inadequate rain data as input to hydrological data is with the assist of Tropical Rainfall Measurement Mission (TRMM) satellite rainfall data which using remote sensing technology (satellite). The purpose of this study is to look for correlations and data corrections and validate TRMM satellite data with rainfall data at the rain station and discharge observation data. Lesti sub-watershed is used as a case study with consideration of the data availability that is considered sufficient. The validation results of corrected TRMM rain data produce Nash-Sutcliffe Efficiency (NSE), Root Mean Squared Error (RMSE), Correlation Coefficient (R), and Relative Error (KR). Then, conducted an analysis of the flow discharge estimation using TRMM rainfall data and validated with Tawangrejeni Automatic Water Level Record (AWLR) data. The results of flow discharge validation using the FJ Mock Method produce an NSE value of 0.507, RMSE 19.383, Correlation Coefficient (R) 0.713, and Relative Error of 0.001. Overall analysis shows TRMM data can be used as an alternative of the rain data that is used to estimate flow discharge, but the result of flow discharge analysis is still better using rainfall data from the rain station post. Keywords: rainfall, TRMM, rain station post, validation, flow discharge, FJ Mock 1. Introduction The rainfall data information is very important for various analyzes of water resources. Rainfall data can be in the form of temporal (time series) or spatial [9]. As one of the important data in hydrological analysis, rainfall data obtained from measurements at the rain station post, so that the rainfall data obtained are expected to have sufficient accuracy. Rainfall data in time series recording can provide trend information from the nature of rain in a place whether it has increased or vice versa. From this description can be said that rainfall data is climatological data that is quite important. Accurate and timely observations and estimates of regional and global precipitation are crucial for various researches and applications [6]. 1 Cite this as: Suryaningtyas, L., Ery, S., & Rispiningtati, R. (2020). Hydrological Analysis of TRMM (Tropical Rainfall Measuring Mission) Data in Lesti Sub Watershed. Civil and Environmental Science Journal, 3(1), pp.18-30. doi: https://doi.org/10.21776/ub.civense.2020.00301.3 Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 19 In fact to obtain the representative rainfall observation data namely both in terms of quality and quantity or length of its observation data that quite appropriate with the requirements is very difficult. The difficulty to get rainfall data, due to the limited number of measuring devices or gauge especially in remote areas, so that it will be difficult to conduct studies and analysis of water resources based on rainfall data in a place because not all places have rainfall monitoring stations manually or automatically [8]. According to Syaifullah [9], the latest technological development, namely in the form of satellite technology (remote sensing) is able to make a breakthrough in terms of the acquisition of rainfall information (precipitation) because with remote sensing technology now has been able to conduct precipitation measurement from remote distance. Areas that do not have sufficient rain recording stations are almost impossible to measure rainfall, but with this technology it is possible to obtain precipitation data that is not limited in space and time, so that it can simply be said that with satellite technology rainfall data can be obtained anytime and anywhere. Tropical Rainfall Measuring Mission (TRMM) satellite has achieved some research progress since its launch in 1997 [1]. TRMM satellite was launched in November 1997 and has been producing since 1998 [3]. One of the satellite technology that has been developed is the TRMM meteorological satellite, which has two types of data namely TRMM NASA (3B42RT) developed by National Aeronautics and Space Administration and TRMM Jaxa (GSMap_NRT) developed by Japan Aerospace Exploration Agency (JAXA), which its results in the form of rainfall data (precipitation) that occurs in the atmosphere with a certain spatial resolution and within period of 3 hours, daily or monthly [9]. Considering that, then it is necessary to conduct evaluation whether the rainfall data from the TRMM satellite and from the existing rain station post network will produce maximum information so that can be obtained the magnitude of rainfall at all points with sufficient accuracy or even differ greatly. In the Lesti Sub Watershed with an area of 381,21 km2 and has five closest rain stations with the uneven position of the rain station post. This study will examine how the correlation of the rainfall station post data towards the satellite rainfall data. This analysis is carried out in an effort to get the value of correlation and accuracy in the results of the analysis of flow discharge estimation using rainfall data at the observation station and satellite rainfall. 2. Materials and Methods 2.1 Materials Data needed for this analysis, namely data of rain station coordinate, DEM, topographic map and river network map, daily rainfall data from the 5 closest rain stations in the Lesti Sub-Watershed for 17 years (2002-2018), TRMM rainfall data per 3 hours (TRMM_3B42RT v7), Tawangrejeni AWLR discharge data for 12 years (2007-2018), and Lesti Sub Watershed land use data. 2.2 Method 2.2.1. Hydrological Analysis Consistency Test Data consistency test is carried out to find out whether there is any deviation in the available rainfall data, so that it can be known whether the data is suitable to be used in further hydrological analysis or not. In this study 2 (two) methods were performed, namely (1) double mass curves; (2) Rescaled Adjusted Partial Sums (RAPS) [7]. Homogeneity Test A series of hydrological data that is presented chronologically as a function of the same time is called a periodic series. The field data that published in general are discharge data, rainfall data, and others. Data is arranged in a series of periodic forms, so that before used for further analysis must be tested. Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 20 The data testing intended are: (1) Test for No Trend; (2) Stationary Test; (3) Persistence Test. The three stages of testing are often referred to as data filtering. 2.2.2 Thiessen Polygon Method Rain station post rainfall data that will be used in the form of regional average rainfall data which calculated using the Thiessen Polygon Method. 2.2.3 TRMM Rainfall Data Validation Test For validation test, using the method of Nash-Sutcliffe Efficiency (NSE), Correlation coefficient (R), Root Mean Squared Error (RMSE), and Relative Error (RE). There are two validation analysis performed, namely validation of uncorrected TRMM data and validation of corrected TRMM data [5]. Validation of uncorrected TRMM data using rain station post rainfall data and uncorrected TRMM. The periods used are monthly with a data length of 7 years (2011-2018), 4 years (2014-2018), 1 year (2018). As for the corrected TRMM data validation conducted a number of processes first, namely calibration, verification, and validation. Calibration and verification using the scatter plot method. For calibration used monthly periods with data length of 10 years (2002-2011), 13 years (2002-2014) and 16 years (2002-2017). While the verification and validation test uses a monthly period with a data length of 7 years (2011-2018), 4 years (2014-2018), 1 year (2018), excluding calibration years. 2.2.4. Analysis of Rain Data into Flow Discharge with F.J. Mock Method Analysis of rainfall data into flow discharge in this study uses the Mock Method which in principle takes into account water balance above the surface and water balance in the ground (groundwater) which is affected by rain, soil type and climate [2, 4]. As for the data used in the flow discharge analysis using the FJ Mock Method, among others: 1. Results of the Regional Average Rainfall Analysis in Lesti Sub Watershed in 2007 - 2008 2. TRMM Rainfall Data before being corrected and after being corrected in 2007-2018 3. Tawangrejeni AWLR discharge recording data for 2007-2018 4. Data on climate recording at Karangploso Climatology Station in 2007-2018, as for the measured data needed are : t = monthly average temperature (oC) RH = monthly average relative humidity (%) n/N = monthly sun brightness (%) u = monthly average wind speed (m/sec) 5. Coordinate data of the observation point namely the point where AWLR Tawangrejeni is located. LL = Latitude Location of location being reviewed 6. The initial storage value is obtained by trial and error 7. The initial groundwater storage value is obtained by trial and error After obtaining the F.J Mock discharge value with the rain station post data, TRMM before and after corrected, the discharge data is analyzed its validation with AWLR discharge data using the method of Nash-Sutcliffe Efficiency (NSE), Correlation Coefficient (R), Root Mean Squared Error (RMSE), and Relative Error (RE). As for the validation method formula used in this study, namely: 1. Nash-Sutcliffe Efficiency (NSE) This method shows how well the plot of the observation value (measurement) is compared to the prediction-simulation value, according to the 1: 1 line, with a range of values ∞ to 1. In other words, the closer to 1, then the better the NSE value. Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 21 NSE = 2 1 2 )( )( 1 ii N i ii XX YX − − −  = (1) With: Xi = observation data (actual data) Yi = estimation data (estimation result data) Xi = average observation data N = the number of data Table 1. Criteria of Nash-Sutcliffe Efficiency (NSE) Value NSE Value Interpretation NSE > 0.75 Good 0.36 < NSE < 0.75 Qualified NSE <0.36 Not Qualified 2. Correlation Coefficient The purpose of this analysis is to obtain a pattern and closeness relationship between two or more variables. R =       = == = = = = −− −− N i N i ii N i N i ii N i N i N i ii YYNXXN YiXYiXN 1 1 22 1 1 22 1 1 1 )()( .......................................... (2) With: Xi = observation data (actual data) Yi = estimation data (estimation result data) N = the number of data Table 2. Criteria of Correlation Coefficient Value R Value Interpretation 0 – 0.19 Very Low 0.20 – 0.39 Low 0.40 – 0.59 Moderate 0.60 – 0.79 Strong 0.8 – 1 Very Strong 3. Root Mean Squared Error (RMSE) RMSE= N YX N i ii = − 1 2 )( (3) With: Xi = observation data (actual data) Yi = estimation data (estimation result data) N = the number of data Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 22 4. Relative Error Test This test is used to determine the comparison between the magnitudes of one variable against other variables used as a benchmark for actual variables. KR = i N i ii Y YX = − 1 )( 100% (4) With: Xi = observation data (actual data) Yi = estimation data (estimation result data) N = the number of data 3. Results and Discussion 3.1. Hydrological Analysis Consistency Test Consistency Test is carried out by two methods, the double mass curve method for station post rainfall data and the RAPS method for TRMM data and discharge data Figure 1 Double Mass Curve of Figure 2 Double Mass Curve of Figure 3 Double Mass Curve of dampit Rain Station Post Poncokusumo Rain Station Post Tumpukrenteng Rain Station Post Figure 4 Double Mass Curve Figure 5 Double Mass Curve of Turen Rain Station Post of Wajak Rain Station Post y = 0.9815x 0 5000 10000 15000 20000 25000 30000 35000 40000 0 10000 20000 30000 40000 C u m u la ti v e o f D a m p it R a in S ta ti o n P o st (m m ) Cumulative of the Surrounding Rain Station Post(mm) y = 0.988x 0 5000 10000 15000 20000 25000 30000 35000 40000 0 10000 20000 30000 40000C u m u la ti v e o f P o n c o k u su m o R a in S ta ti o n P o st (m m ) Cumulative of the Surrounding Rain Station Post(mm) 44.47O 44.65 O y = 0.928x 0 5000 10000 15000 20000 25000 30000 35000 40000 0 10000 20000 30000 40000 C u m u la ti v e o f T u m p u k re n te n g R a in S ta ti o n P o st (m m ) Cumulative of the Surrounding Rain Station Post(mm) 42.86O y = 0.9978x 0 5000 10000 15000 20000 25000 30000 35000 40000 0 10000 20000 30000 40000 C u m u la ti v e o f T u re n R a in S ta ti o n P o st (m m ) Cumulative of the Surrounding Rain Station Post(mm) 44.94O y = 1.1076x 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 0 10000 20000 30000 40000C u m u la ti v e o f W a ja k R a in S ta ti o n P o st (m m ) Cumulative of the Surrounding Rain Station Post(mm) 47.92O Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 23 Table 3 Recapitulation of the α value at each rain station post No Rain Station Post α value 1 Dampit 44.47° 2 Poncokusumo 44.65° 3 Tumpukrenteng 42.86° 4 Turen 44.94° 5 Wajak 47.92° Table 4 Recapitulation of Consistency Test Results No Post Name Double Mass Curve Method RAPS Method Information Angle Q/n0,5 calculate Q/n0,5 table R/n0,5 calculate R/n0,5 table 1 Dampit 44.47° - - - - Consistent 2 Poncokusumo 44.65° - - - - Consistent 3 Tumpukrenteng 42.86° - - - - Consistent 4 Turen 44.94° - - - - Consistent 5 Wajak 47.92° - - - - Consistent 6 AWLR - 0.54 1.16 0.63 1.31 Consistent 7 TRMM - 0.41 1.20 0.54 1.39 Consistent Based on Figure 1 to Figure 5 and Table 3 then can be said that rainfall data of rain station post that used after being tested using the Double Mass Curve Method is consistent because the resulting angle is in the value ranges of 42o< α < 48 o. Whereas based on Table 4, the TRMM rainfall data consistency test and the discharge data using the RAPS Method also meet the test requirements because the value of Qcalculate< Qcritical and the value of Rcalculate< Rcritical so that the results can be considered to be consistent. These test results indicate that the selected data can be used for further hydrological testing and analysis. Homogeneity Test In this study, annual rainfall data of the rainfall station was tested for absence of trends by the Spearman Method using 2-side T-Test. The recapitulation of the test results presented as follows. Table 5 Recapitulation of Test for No Trend Results No. Name of Rain Station Post Tcalculate α tc Information 1 Dampit 0.774 5% 2.131 not indicate a trend 2 Poncokusumo 0.754 5% 2.131 not indicate a trend 3 Tumpukrenteng 0.876 5% 2.131 not indicate a trend 4 Turen 0.266 5% 2.131 not indicate a trend 5 Wajak 2.052 5% 2.131 not indicate a trend 6 AWLR Tawangrejeni 2.561 1% 3.169 not indicate a trend 7 TRMM 0.324 5% 2.131 not indicate a trend Based on Table 5 can be seen that the entire data (except AWLR discharge data) did not indicate a trend by showing tcalculate < ttable at a 5% confidence degree. Thus, these data can be further analyzed. Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 24 Table 6 Recapitulation of Variance Stability Test Results (F Test) No. Name of Rain Station Post Fcalculate α Fc Information 1 Dampit 0.759 5% 3.370 The value of the variance is stable 2 Poncokusumo 1.168 5% 3.370 The value of the variance is stable 3 Tumpukrenteng 5.658 1% 6.840 The value of the variance is stable 4 Turen 1.449 5% 3.370 The value of the variance is stable 5 Wajak 0.830 5% 3.370 The value of the variance is stable 6 AWLR Tawangrejeni 22.326 1% 10.97 The value of the variance is not stable 7 TRMM 1.416 5% 3.370 The value of the variance is stable Table 7 Recapitulation of Average Stability Test Results (t Test) Annual Period No. Name of Rain Station Post Tcalculate α tc Information 1 Dampit 0.856 5% 2.131 The average value is stable 2 Poncokusumo 1.190 5% 2.131 The average value is stable 3 Tumpukrenteng 0.617 5% 2.131 The average value is stable 4 Turen 0.727 5% 2.131 The average value is stable 5 Wajak 1.183 5% 2.131 The average value is stable 6 AWLR Tawangrejeni 1.906 5% 2.228 The average value is stable 7 TRMM -0.129 5% 2.131 The average value is stable From Table 6 and Table 7 above it can be seen that the value of F calculate < the value of F table and the value of t calculate < the value of t table, so it can be concluded that the rainfall data of the five rain station posts, the TRMM rainfall data, and the discharge data used have a stable variance and average. The persistence test is an independent test for each value in the periodic series. First, the number of serial correlation coefficients must be calculated by the Spearman Method, then the calculation of the persistence test with the T-Test is conducted. The recapitulation of the test results is presented as follows. Table 8 Recapitulation of Persistence Test Results No. Name of Rain Station Post Tcalculate α tc Information 1 Dampit -1.65 5% 2.145 Data is random 2 Turen -1.432 5% 2.145 Data is random 3 Tumpakrenteng -2.028 5% 2.145 Data is random 4 Wajak -3.840 5% 2.145 Data is random 5 Poncokusumo -0.828 5% 2.145 Data is random 4 Tawangrenjani AWLR Discharge -0.178 5% 2.262 Data is random 5 TRMM -0.793 5% 2.145 Data is random Based on Table 8 it can be seen that almost all of the data are random by showing tcalculate< ttable at 5% confidence level/degree. Thus, these data can be analyzed further. 3.2 Correlation of Rain Data of Rain Station Post and TRMM Based on Table 9, the correlation analysis results of all rain station posts with TRMM data (2002- 2018 data), have a good correlation with TRMM rain data, this can be seen from the correlation coefficient values that are at values> 0,6. Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 25 Table 9 Correlation Results of Monthly Rain Data of Rain Post with TRMM No. Post Correlation 1 Dampit 0.81 2 Poncokusumo 0.81 3 Tumpuk Renteng 0.78 4 Turen 0.86 5 Wajak 0.85 Table 10 Thiessen coefficient of Lesti Sub Watershed No. Post Area (km2) Kr 1 Dampit 200.601 0.659 2 Poncokusumo 89.476 0.294 3 Tumpuk Renteng 1.299 0.004 4 Turen 12.937 0.043 5 Wajak 76.894 0.253 Total 304.313 1 3.3 Regional Average Rainfall Analysis Figure 6 Map of the Influence Area of the Lesti Sub Watershed Rain Station Post by using the Thiessen Polygon Method Based on Table 10 obtained that the results of the value of Kr for each post of the rain station is a comparison of the area of influence of each post of the rain station on the area of the Lesti Sub Watershed towards the total area of the Lesti Sub Watershed. The calculation results of the Kr value are then used to calculate the regional average rainfall. Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 26 Figure 7 shows a comparison graph of regional average rainfall with TRMM before being corrected. From the graph it can be seen that the TRMM rainfall value tends to be smaller, but has a pattern and fluctuation that is almost the same as the regional average rainfall. Further analysis after obtaining regional average rainfall is a validation analysis of rainfall data from the rain station post and TRMM rainfall. Figure 7 Graph of comparison of average regional rainfall with TRMM 3.4 Calibration and Validation Analysis of Rain Data from Rain Station Post and TRMM TRMM Rain Data Calibration Figure 8 through Figure 13 shows the calibration scatterplot to get the best equation. From the regression equation that has been obtained to get the corrected TRMM rain data then the used regression equation with the largest R2 value. Obtained the results of the TRMM rainfall regression equation in the Lesti watershed with R² = 0.7158 with the polynomial equation. Figure 8 Linear Regression Equation Figure 9 Intercept Linear Regression Equation Figure 10 Polynomial Regression Figure 11 Rank Regression Equation Equation y = 1,0837x + 53,792 R² = 0,7049 0 200 400 600 800 1000 0 200 400 600 800 R a in S ta ti o n P o st R a in fa ll ( m m ) Monthly TRMM Rainfall(mm) Data Calibration of Station Post and TRMM 16 Years (2002-2017) Linear Function y = 1,2626x R² = 0,6743 0 200 400 600 800 1000 0 200 400 600 800R a in S ta ti o n P o st R a in fa ll ( m m ) Monthly TRMM Rainfall(mm) Data Calibration of Station Post and TRMM 16 Years (2002-2017) Intercept Linear Function y = -0,0007x2 + 1,4289x + 30,867 R² = 0,7158 0 200 400 600 800 1000 0 200 400 600 800R a in S ta ti o n P o st R a in fa ll (m m ) Monthly TRMM Rainfall(mm) Data Calibration of Station Post and TRMM 16 Years (2002-2017) Polynomial Function y = 3,0693x0,8318 R² = 0,6561 0 200 400 600 800 1000 0 200 400 600 800 R a in S ta ti o n P o st R a in fa ll ( m m ) Monthly TRMM Rainfall(mm) Data Calibration of Station Post and TRMM 16 Years (2002-2017) Rank Functiont Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 27 Figure 12 Logarithmic Regression Figure 13 Intercept Polynomial Regression Equation Equation TRMM Rain Data Verification The next stage is verification of the data outside the data used for calibration. Figure 14 Verification of TRMM Rainfall in 2018 Figure 14 shows a verification graph of rain data of rain station post with TRMM outside the calibration year. The correlation value (R) produced for the 2018 regional rainfall is 95,98%, this shows that the corrected TRMM rainfall data has a very strong correlation with station post rainfall data. TRMM Rain Data Validation Validation is performed on data outside the data used for calibration. To be able to measure the magnitude of the difference in the results of the model calculation towards the observational data then conducted TRMM rain data validation using the objective function of the NSE (Nash-Sutcliffe Efficiency), Correlation Coefficient (R), RMSE (Root Mean Square Error), and Relative Error (RE). Table 11 Recapitulation of Calculation Results for Validation of Station Post Rainfall Data with TRMM Before Corrected Total Year NSE RMSE KR R Calibration Validation Value Interpretation Value Interpretation 10 7 0.545 Qualified 114.949 0.269 0.829 Strong 13 4 0.476 Qualified 122.960 0.282 0.807 Moderate 16 1 0.631 Qualified 108.479 0.416 0.959 Very Strong R = 0.9598 0 50 100 150 200 250 300 350 400 450 500 0 200 400 600R a in S ta ti o n P o st R a in fa ll ( m m ) Monthly TRMM Rainfall(mm) Data Verification of Station Post and TRMM 1 Year (2018) y = 90,312ln(x) - 157,15 R² = 0,547 -400 -200 0 200 400 600 800 1000 0 200 400 600 800 R a in S ta ti o n P o st R a in fa ll ( m m ) Monthly TRMM Rainfall(mm) Data Calibration of Station Post and TRMM 16 Years (2002-2017) Logarithmic Function y = -0,0011x2 + 1,6534x R² = 0,7092 0 200 400 600 800 1000 0 200 400 600 800 R a in S ta ti o n P o st R a in fa ll ( m m ) Monthly TRMM Rainfall(mm) Data Calibration of Station Post and TRMM 16 Years (2002-2017) Intercept Polynomial Function Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 28 Table 12 Recapitulation of Calculation Results for Validation of Station Post Rainfall Data with TRMM After Corrected Total Year NSE RMSE KR R Calibration Validation Value Interpretation Value Interpretation 10 7 0.657 Qualified 99.726 0.045 0.835 Strong 13 4 0.674 Qualified 97.033 0.007 0.825 Medium 16 1 0.890 Good 59.163 0.116 0.960 Very Strong Figure 15 Graph of Lesti Watershed Rainfall in 2007-2018 Based on Table 11 and Table 12 by comparing the validation results of uncorrected data and corrected data, the corrected data validation results have better results. The results of the data validation show the results "Qualified" at the NSE method for all data, except for 1 year validation of the corrected data obtained the results of "Good". The RMSE value for uncorrected data validation is relatively high, but has decreased in the validation of corrected data. The value of Relative Error is classified as very small in all data. For the Correlation Coefficient of all data, the calculation results show a very strong relationship, but the best results are found in the validation of 1-year corrected data using 16-year data calibration. This shows that the more data used for calibration, then the better it is for validation. 3.5 Water Discharge Analysis FJ Mock Methode Flow discharge simulation is carried out by trial and error of parameter values which are carried out repeatedly until it meets the model performance criteria. Based on the calibration results obtained, the simulation discharge hydrograph approaches the observation discharge. The trial results of the parameter values can be seen in Table 13. Figure 16 Graph of Model Discharge and Observation Discharge of Lesti Watershed in 2002-2018 0,0 100,0 200,0 300,0 400,0 500,0 600,0 700,0 800,0 900,0 1000,0 Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018R iv e r F lo w D is c h a rg e ( m ³/ se c ) Month Rainfall of Lesti Watershed in 2007 - 2018 0,0 20,0 40,0 60,0 80,0 100,0 120,0 140,0 160,0 180,0 Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t Ja n A p r Ju l O c t 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 R iv e r F lo w D is c h a rg e ( m ³s e c ) Month Calculation Discharge and Observation Discharge in 2007-2018 Debit AWLR Debit CH Wilayah Debit TRMM Tidak Terkoreksi Debit TRMM Terkoreksi Civil and Environmental Science Journal Vol. III, No. 01, pp. 018-030, 2020 ol. xx, No. xx, pp. xxx-xxx, 20xx 29 Table 13 Simulation Results for FJ Mock Parameters No Year Parameter SMC (mm) i k IS (mm) Vn (mm) 1 2007 250 0.53 0.09 50 50 2 2008 250 0.21 0.92 50 50 3 2009 250 0.08 0.10 50 50 4 2010 250 0.72 0.99 50 50 5 2011 250 0.61 0.97 50 50 6 2012 250 0.56 0.99 50 50 7 2013 250 0.97 0.97 50 50 8 2014 250 0.79 0.99 50 50 9 2015 250 0.91 0.99 50 50 10 2016 250 0.95 0.99 50 50 11 2017 250 0.89 0.99 50 50 12 2018 250 0.98 0.99 50 50 From Figure 16 it can be seen that the trend of monthly AWLR discharge and the FJ Mock Method flow discharge have almost the same pattern. From the table, it appears that the AWLR data for 2008- 2009 is indeed that the value is very far from the average each year which occurs every season. Validation of F.J. Mock Flow Discharge To show the accuracy value of each flow discharge simulation from the different rain database with the FJ Mock Method, conducted validation test of observation Mock discharge data (AWLR) using the objective function of NSE, Correlation Coefficient, RMSE and Relative Error. Table 14 Validation Results on FJ Mock Flow Discharge No Discharge of Model Results with database Value NSE R RMSE KR 1 Rain Station Post Rainfall 0.507 0.713 19.383 0.001 2 TRMM Rainfall 0.374 0.614 21.839 0.016 3 Corrected TRMM Rainfall 0.411 0.646 21.190 0.025 Based on Table 14 above, the NSE value of the model discharge with three variations of rain data has a value> 0.36 which indicates that the discharge analysis of the model result can be said qualified. But the NSE value of discharge data with TRMM database has the smallest value, so that it is better to use corrected TRMM data. While the value of the correlation coefficient (R) from the three sources of rain data has a value of 0,5