Data and Sources Properties Required Tools Conclusion

III. RESEARCH METHODOLOGY

3.1 Time and Location

The research was conducted from February until June 2007 at MIT- BIOTROP covering Karawang District, West Java. Image processing and analysis of satellite data have been conducted from March to May 2007 at the National Institute of Aeronautics and Space LAPAN.

3.2 Data and Sources Properties

1. Remote sensing data Landsat-7 ETM+ obtained from LAPAN between 2003 until 2006 series data per year for wet or dry session. 2. Vectors data administration from Bakosurtanal and irrigation network from Local Government and DEM-SRTM from USGS. 3. Climatic data from BMG series data rainfall 1971-2000, air temperature and wind velocity 1995 – 2002 and from PJT-II Jatiluhur series data rainfall 2004-2006. 4. Field data coordinate position, land cover and land use checking from classification Landsat ETM+ between actual conditions in the field. 5. Statistical data drought vulnerability, paddy field, discharge and volume of water.

3.3 Required Tools

The hardware tools used for this research consist of Personal Computer Pentium IV with Processor 1.8 GB, RAM 526 MB, Hard disk 60 GB, LaserJet 2500 Printer, and hand held Global Positioning System GPS device, while the softwares consist of ER Mapper 7, Arc View 3.3, and Microsoft Office 2003. 25 The softwares used to publish information through Web GIS are: - MS4W Map Server for Windows. MS4W, a free software package, that is Apache for web server application i.e., Map Server; this software is used to develop web based GIS application. With the use of PHP script, map server script can process the spatial database and non-spatial database - MySQL 4.0.18 is a software used to store and manage all related data. This is a free software for database application - Macromedia Dreamweaver MX 2004 is used to develop web pages.

3.4 Methodology

Data flow diagram of this research methodology is described in Figure 3. The following individual components will be explained in detail. Map Data RS Data Field Data Climatic Data Selecting and Cropping data Radiometric Geometric Correction NDVI Albedo Surface Temperature Land Cover Classification Emisivity Constantan Short Radiation Wind Velocity Air Temperature Extraterrestrial Radiation Topography, Irrigation, Paddy Field DEM Interpolation Energy Balance Model Evaporative Fraction EF Map Overlay Spatial Information of Water Stress for Drought Indicator Statistic Data Drought Location Paddy Field Debit Volume Water Web GIS INTERNET Data Collection Map Data RS Data Field Data Climatic Data Selecting and Cropping data Radiometric Geometric Correction NDVI Albedo Surface Temperature Land Cover Classification Emisivity Constantan Short Radiation Wind Velocity Air Temperature Extraterrestrial Radiation Topography, Irrigation, Paddy Field DEM Interpolation Energy Balance Model Evaporative Fraction EF Map Overlay Spatial Information of Water Stress for Drought Indicator Statistic Data Drought Location Paddy Field Debit Volume Water Web GIS INTERNET Data Collection Figure 3. Flow diagram of research methodology 26

3.4.1 Image Pre-Processing

3.4.1.1 Selecting and Cropping Data

The aim of selecting data is to get the free cloud-cover image data. Furthermore, to get the efficiency in image processing, the selected data are cropped for the study area.

3.4.1.2 Radiometric and Geometric Correction

Radiometric and geometric errors are the most common types of error encountered in remotely sensed imagery. To remove or minimize the error the following should be done: 1. Radiometric correction, using histogram adjustment 2. Geometric correction is the basic operation in geometric rectification which covers collecting ground control point GCP, transformation and resampling. The topographic map from Bakosurtanal of Karawang district with the scale of 1:25.000 is used to correct the image data.

3.4.2 Data Processing and Analysis

Satellite imagery data is analyzed using ER Mapper 7 and vector data is processed using ArcView 3.3. Field data are collected from surveys and position of the location is taken by using GPS. Field data is used to verify result of land cover. Result analysis is thematic images of evaporative fraction from 2003 until 2006.

3.4.3 Energy Balance Modeling

Energy balance estimation has been performed using daily meteorological data from surface station and daily remote sensing data from Landsat. The flow diagram modeling of energy balance is presented in Figure 4. 27 Remote Sensing Data Climate Data NDVI Ts α Ta U 2 Rs Net Radiation Rn 4 16 . 273 1 + − + − = Ts Rl Rs Rn εσ α Net Radiation Rn 4 16 . 273 1 + − + − = Ts Rl Rs Rn εσ α Sensible Heat Flux H 273 900 2 Ta Ts U Ta H − + = λ γ Sensible Heat Flux H 273 900 2 Ta Ts U Ta H − + = λ γ Soil Heat Flux G 0.98NDVI - 1 007 . 0.0038 4 2 α α α + = Ts Rn G Soil Heat Flux G 0.98NDVI - 1 007 . 0.0038 4 2 α α α + = Ts Rn G Energy Evapotranspiration λE H G Rn E − − = λ Energy Evapotranspiration λE H G Rn E − − = λ Evaporative Fraction EF G Rn E EF − = λ Evaporative Fraction EF G Rn E EF − = λ Figure 4. Flow diagram of energy balance modeling

3.4.4 Development of Web GIS

In this research we also develop a web GIS to publish the spatial information of energy balance Figure 5. Developing web GIS is divided into three major steps. First, is the design data information of drought indicator. Second, is building web page based on drought information. The last step is publishing the spatial drought information through the internet. Figure 5. Schema to build and publish information 28 The raw materials for information system are data input, i.e. the representation of facts or concepts that becomes usable when processed. Data is required to produce information which is needed by users. There are conceptual, logical and physical database designs. Conceptual design focuses on what data should be stored in the database, while the function model deals with how the data is processed. To put this in the context of the relational database, the data model is used to design the relational tables. The functional model is used to design the queries that will access and perform operations on those tables. Logical database design focuses on constructing a model of information used in an enterprise based on specific data model. Logical model explains the logic of database which is the process of constructing a model of information used in an enterprise based on specific data model, but independent of particular DBMS and other physical considerations. Physical design describes the physic of database visualization, which allows the designer to make decisions on how the database is to be implemented. Physical model describes the data physically where entity is set based on database software. This phase building web page also applies to the programs which process the database. In the system development approach, the parallel activities could be observed. The designed web page of the system is divided into two parts. 1. Design of spatial modeling includes spatial data selection and spatial data manipulation. 2. Design of output includes map, graph and table. The last step is implementation to improve the system. A standard web browser will be used as the web page. The browser connects to the server, which contains 29 spatial database. The server sends back a result of request to the client. An interface will be developed which allows user to query the spatial database. The result is returned to the user’s browser as an HTML file that contains the dataset in image format usually .pnggif format. 30 IV. RESULTS AND DISCUSSION

4.1. Energy Balance

The component is analyzed from Landsat-7 and climate data. Landsat-7 ETM+ image data used is from 2003 to 2006. Mosaic process is applied only to data 2006 to see land cover condition. The data actually was invalid since 2003 and a lot of striping data cannot be analyzed. Climate data are obtained from BMG air temperature and wind velocity data acquisition 1995 to 2002 and from PJT-II Jatiluhur rainfall data acquisition 2004 to 2006. 4.1.1. Existing Land Use The exiting land use map is obtained by interpreting Landsat-7 ETM+ image in October 2006. The land use map consists of paddy field, industrial area, residential area, fish pond, vegetation cover, unirrigated land and water body. Table 5 shows Karawang district where 22 of the total area is paddy field area and 1.8 industrial area. Because in October 2006 usually start of the rainy season, but there was no rainfall observed. Based on data form BPS Karawang total area paddy field at 2006 is 82,285 ha. So existing paddy field detected is low and partly detected is unirrigated land. Data used is Landsat-7 ETM+ after mosaic process. Table 5. The existing land use of Karawang district No Land Use Area ha Area 1 Paddy Field 42,424 22 2 Industry 3,476 2 3 Residential 10,282 5 4 Fish Pond 14,777 8 5 Vegetation 51,889 27 6 Unirrigated Land 69,891 36 7 Water 357 Total 193,097 100

4.1.2. Albedo

The albedo of an object is the extent to which it reflects light, defined as ratio of reflected to incident electromagnetic radiation. Its value depends on the frequency of radiation unqualified, it usually refers to same appropriate average across the spectrum of visible light. In general, the albedo depends on the 31 direction and direction distribution of incoming radiation. Around 20,000-30,000 ha area cannot be analyzed due to a striping data since 2003. Acquisition: 632003 Acquisition: 4182004 Acquisition: 652004 Acquisition: 7232004 Acquisition: 4212005 Acquisition: 7102005 Acquisition: 8112005 Acquisition: 1012006 Legend Figure 6. Distribution of albedo in Karawang district 32 Table 6. Albedo from Karawang district 632003 4182004 652004 7232004 4212005 7102005 8112005 1012006 0-0.1 22,364 19,029 39,652 28,152 25,128 46,825 15,435 12,972 0.1-0.2 149,146 149,049 110,634 135,556 115,995 116,127 150,699 156,612 0.2-0.3 583 1,035 6,623 6,341 13,220 1,137 1,711 2,581 0.3-0.4 74 287 3,150 337 4,958 320 357 49 0.4-0.5 49 262 3,826 47 4,198 164 312 6 0.5-0.6 3 48 2,811 1,255 29 171 0.6-0.7 519 16 0.7-0.8 0.8-0.9 0.9-1 No Data 21,022 23,533 26,026 22,808 28,486 28,639 24,540 21,022 Total 193,241 193,241 193,241 193,241 193,241 193,241 193,241 193,241 Albedo Date Area ha Table 6 shows area of albedo range in Karawang district, and the distribution of albedo. The dominant albedo is 0.1-0.2, the indicator area are vegetation, residential, industrial, and unirrigated lands. Albedo with range of 0- 0.1 is dominant in water body and fish pond areas. It indicaties that a lot of radiant reflected from surface is low and radiation absorbed is high.

4.1.3. Surface Temperature

Land surface temperature is one of the key parameters of land surface process, combining surface-atmosphere interaction and energy fluxes between the atmosphere and the ground. Table 7 shows the distribution of surface temperature at Karawang district. The value of land surface temperature ranges from 24-34 o C detected from thermal sensor of Landsat-7 ETM+ satellite. Based on Figure 7 the surface temperature value of October 01, 2006 is high. Table 7. Surface temperature in Karawang district. 632003 4182004 652004 7232004 4212005 7102005 8112005 1012006 24-25 102 4,782 24,432 10,941 1,135 289 2 25-26 10 317 5,443 11,181 8,533 4,587 515 7 26-27 205 1,885 6,325 17,096 36,622 16,748 1,086 32 27-28 44,197 64,421 42,995 46,838 102,035 125,211 83,370 2,539 28-29 94,448 85,636 92,316 52,874 6,255 14,563 67,609 25,529 29-30 32,642 16,389 15,119 16,530 360 2,328 15,516 42,425 30-31 715 923 235 1,464 9 30 316 40,766 31-32 2 25 1 19 1 1 40,497 32-33 12, 33-34 8, No Data 21,022 23,543 26,026 22,808 28,486 28,639 24,540 21,022 Total 193,241 193,231 193,241 193,241 193,241 193,241 193,241 193,241 TS Date Area ha 406 016 33 Acquisition: 632003 Acquisition: 4182004 Acquisition: 652004 Acquisition: 7232004 Acquisition: 4212005 Acquisition: 7102005 Acquisition: 8112005 Acquisition: 1012006 Legend o C Figure 7. Distribution of surface temperature in Karawang district Distribution pattern of surface temperature shows higher surface temperature in residential and industrial areas than other areas, for example Landsat data of August 11, 2005. Surface temperatures in residential and industrial areas range between 29-31 o C, while the other land cover classifications are lower. Very significant condition is shown in October 1, 2006, where the land 34 surface is relatively dry because there was no rain for a successive time and limited distribution of water from Jatiluhur dam, consequently the surface temperature of non-vegetation is higher 29-34 o C , while for vegetation area it is less than 29 o C. 4.1.4. NDVI Based on Figure 8 the vegetation index number in April 2004 is higher than in June-July 2004, while the condition in 2005 showed that for the wet season it is lower than dry season July-August. Furthermore, in 2003 the vegetation index is moderate especially in paddy field areas, while in 2006 the distribution is different with low vegetation index or mostly negative. Vegetation index area Normalized Different Vegetation Index is dominant at moderate level 0-0.6 except in October 2006 where the dominance is at low level -0.6-0. Table 8. NDVI in Karawang district. 632003 4182004 652004 7232004 4212005 7102005 8112005 1012006 -1--0.8 5 1 33 12 -0.8--0.6 123 131 93 60 57 56 4 119 -0.6--0.4 9,545 6,881 8,884 10,385 6,716 7,044 5,215 5,208 -0.4--0.2 41,670 31,168 48,430 59,101 53,720 44,969 36,733 52,157 -0.2-0 42,765 38,734 46,870 40,907 59,219 37,240 48,599 70,431 0-0.2 32,341 37,598 41,678 34,672 29,609 30,227 45,068 30,470 0.2-0.4 42,600 53,387 19,602 24,113 14,692 39,125 32,745 13,427 0.4-0.6 2,730 1,591 1,149 924 190 5,478 1 393 0.6-0.8 0.8-1 214 508 269 518 452 336 11 No Data 21,467 23,533 26,026 22,808 28,486 28,639 24,540 21,022 Total 193,241 193,241 193,241 193,241 193,241 193,241 193,241 193,241 NDVI Date Area ha 1 3 Acquisition: 632003 Acquisition: 4182004 Acquisition: 652004 35 Acquisition: 7232004 Acquisition: 4212005 Acquisition: 7102005 Acquisition: 8112005 Acquisition: 1012006 Legend Figure 8. Distribution of NDVI in Karawang district.

4.1.5. Wind Velocity

Wind velocity is obtained from West Java and Jakarta areas using spatial interpolation kriging. The wind velocity data is obtained from 10 stations surrounding Karawang district. Temporal resolution data is monthly average from 1995 to 2002. Wind velocity variation from January until December indicates higher value in northern side of Karawang especially coastal area than southern side. Based on Beaufort scale from January to December wind velocity can be classified into high, moderate, and low wind velocities. High category is wind velocity more than 6 ms, moderate 3-6 ms and low, less than 3 ms. High category is dominant in northern side, moderate in center Karawang and low in southern area based on from topographic condition of Karawang district. 36 Southern area of Karawang district is high land with air pressure higher than the northern area. January February March April May June July August September 37 October November December Legend Figure 9. Distribution of wind velocity ms in Karawang district.

4.1.6. Air Temperature

Air temperature obtained by interpolation method from data collected around Karawang district shows different pattern and class for each month. Average temperature in February is less than 25 o C. This maybe due to the high temperature of Karawang district during the rainy season caused by very low radiation to land surface and much radiation is emitted by the clouds. Trend of temperature from January until December is relative similar, but temperature value is different. Northern area of Karawang is higher than southern area. The topography condition of Karawang determines temperature degree. North area land altitude is less than 25 meter, while south area is higher. In general, air temperatures in Karawang are around 23-28 o C. 38 January February March April May June July August September 39 October November December Legend Figure 10. Distribution of air temperature o C in Karawang district.

4.1.7. Solar Radiation

Solar radiation Rs is the amount of radiation reaching a horizontal plane. Solar radiation is also referred to as shortwave radiation. For a cloudless day, Rs is roughly 75 of extra terrestrial radiation. On a cloudy day, the radiation is scattered in the atmosphere, but even with extremely dense cloud cover, about 25 of the extra terrestrial radiation may still reach the earths surface mainly as diffuse sky radiation. Extra terrestrial radiation Ra is the solar radiation received at the top of the earths atmosphere on a horizontal surface. If the sun is directly overhead, the angle of incidence is zero and the extraterrestrial radiation is 0.0820 MJ m -2 min -1 . Solar radiation is also known as global radiation, meaning that it is the sum of direct shortwave radiation from the sun and diffuse sky radiation from all upward angles. 40

4.1.8. Net Radiation

The distribution net radiation showing on Figure 11 as follows: Acquisition: 632003 Acquisition: 4182004 Acquisition: 652004 Acquisition: 7232004 Acquisition: 4212005 Acquisition: 7102005 Acquisition: 8112005 Acquisition: 1012006 Legend Wm -2 Figure 11. Distribution of net radiation Wm -2 in Karawang district 41 Net radiation is calculated based on position of latitude and time of Julian day. Total net radiation is shown in Table 9. Table 9. Total net radiation in Karawang district 632003 4182004 652004 7232004 4212005 7102005 8112005 1012006 0-50 50-100 156 100-150 23 40 6,014 7 1,187 121 273 150-200 181 415 6,923 676 7,719 739 658 6 200-250 160,020 2,876 146,872 144,047 34,338 155,413 119,073 257 250-300 11,995 169,079 12,893 28,689 131,469 15,098 52,908 187,332 300-350 1,167 1,035 5,582 350-400 No Data 21,022 19,664 20,382 19,822 17,493 21,871 20,329 63 Total 193,241 193,241 193,241 193,241 193,241 193,241 193,241 193,241 Rn Date Area ha In April Karawang district has net radiation between 250-300 Wm -2 , in May-August 200-250 Wm -2 and October 250-300 Wm -2 . In October almost all area receive net radiation around 250-300 Wm -2 , or 187,000 hectare from total area of Karawang 193000 hectare. Geographically in Karawang district the influence on amount of the net radiation which reaches the land surface is not significant.

4.1.9. Soil Heat Flux

Soil heat flux distribution is shown in Figure 12. Based on Table 10 it is found that in Karawang district potential energy for soil heating is 30-40 Wm -2 , except for the data of October 01, 2006 when total energy used to heat soil is about 40-50 Wm -2 for areas near water body or fish pond and for water body and fish pond it is about 30-40 Wm -2 . Table 10. Area of energy for soil heat flux 632003 4182004 652004 7232004 4212005 7102005 8112005 1012006 0-10 445 437 551 496 551 544 501 10 10-20 84 1,784 2 20-30 12,385 333 29,965 47,818 9,181 71,299 1,085 3 30-40 159,389 172,806 142,259 123,321 166,016 99,527 171,327 65,900 40-50 1 127,263 50-60 No Data 21,022 19,664 20,382 19,822 17,493 21,871 20,329 63 Total 193,241 193,241 193,241 193,241 193,241 193,241 193,241 193,241 Soil Date Area ha 42 Acquisition: 632003 Acquisition: 4182004 Acquisition: 652004 Acquisition: 7232004 Acquisition: 4212005 Acquisition: 7102005 Acquisition: 8112005 Acquisition: 1012006 Legend Wm -2 Figure 12. Distribution energy for soil heat flux Wm -2 in Karawang district

4.1.10. Sensible Heat Flux

Distribution energy used to heat air or sensible heat flux is shown in Figure 13. Sensible heat flux ranges from 50-100 Wm -2 and for several areas are 43 about 100-150 Wm -2 . Except data in October 01, 2006 where total energy used is higher than the other data about 100-200 Wm -2 . The dry months before October 2006 is long i.e. about 4-5 months, therefore the total energy used to heat air is high. Acquisition: 632003 Acquisition: 4182004 Acquisition: 652004 Acquisition: 7232004 Acquisition: 4212005 Acquisition: 7102005 Acquisition: 8112005 Acquisition: 1012006 Legend Wm -2 Figure 13. Distribution energy for sensible heat flux in Karawang district 44

4.1.11. Evapotranspiration

The total energy used for evapotranspiration ranged from 100-200 Wm -2 . Figure 14 presents distribution of energy use for evapotranspiration in Karawang district for vegetation or non vegetation cover. Based on October 2006 data, the distribution is different from other data, because part of Karawang area uses energy about 50-100 Wm -2 . It shows that amount of water content in soil or vegetation is low than total energy needed for evaporation. Land cover such as water body, fish pond and cloud needs higher energy about 200-250 Wm -2 . It indicates that an area covers with water or cloud needs more energy for evaporation. Furthermore, it will decrease total energy used for soil and air heating. Acquisition: 632003 Acquisition: 4182004 Acquisition: 652004 Acquisition: 7232004 Acquisition: 4212005 Acquisition: 7102005 45 Acquisition: 8112005 Acquisition: 1012006 Legend Wm -2 Figure 14. Distribution energy for evapotranspiration in Karawang district

4.1.12. Evaporative Fraction

Evaporative fraction constitutes index value from ratio amount of energy used for evapotranspiration with amount of net radiation minus soil heat flux. EF value is near or equal one indicating this area is not potential of drought. It is caused by amount of energy focused to evapotranspiration process like transpiration and evaporation. The water balance in soil and vegetation is sufficient for transpiration process of crops and evaporation of the land. This drought indication occurs if amount of energy used for evapotranspiration is low, because soil and vegetation water content is little. EF value is near or equal zero indicating this area is potential of drought. Table 11. Total area of evaporative fraction EF in Karawang district 632003 4182004 652004 7232004 4212005 7102005 8112005 1012006 0-0.1 2 9 2 2,980 0.1-0.2 7 6 2 107 8 12,371 0.2-0.3 103 120 54 744 6 32 114 26,028 0.3-0.4 2,636 2,423 1,018 2,105 18 290 1,105 35,673 0.4-0.5 23,141 16,663 8,322 5,404 197 1,725 5,842 35,257 0.5-0.6 46,567 72,356 48,117 16,599 6,880 4,687 16,207 29,567 0.6-0.7 57,529 62,839 61,347 30,174 56,934 10,089 37,320 26,077 0.7-0.8 34,590 17,777 36,808 28,655 77,267 35,343 54,107 19,811 0.8-0.9 7,571 1,231 10,130 25,554 18,953 67,100 51,531 5,180 0.9-1.0 75 7,062 64,069 15,493 52,103 6,677 233 No Data 21,022 19,825 20,382 19,822 17,493 21,871 20,329 63 Total 193,241 193,241 193,241 193,241 193,241 193,241 193,241 193,241 EF Date Area ha 46 Acquisition: 632003 Acquisition: 4182004 Acquisition: 652004 Acquisition: 7232004 Acquisition: 4212005 Acquisition: 7102005 Acquisition: 8112005 Acquisition: 1012006 Legend Figure 15. Distribution of evaporative fraction EF in Karawang district Table 11 and Figure 15 show the distribution of evaporative fraction. The dry or wet seasons in 2003-2005 have low potential of drought, because there is enough water for evapotranspiration. Water source can be supplied by rain and 47 irrigation. On the contrary in 2006 there is a potential for drought in several locations of Karawang.

4.1.13. Irrigation

The source for water irrigation from PJT II Jatiluhur is divided into 3 locations that is Curug dam for southern and eastern areas and Walahar dam for northern area. Northern area receives the much water supply from irrigation. The water debit in Walahar dam during period January-September 2006 is shown in Table 12 and for irrigation network is shown in Figure 16. Table 12. Discharge of water in Walahar dam m 3 s and Volume m 3 Period Water Use Q1 Available Water Q2 Water volume of Q1 Water volume of Q2 1-15 January 90.48 257.10 117,268,300.80 333,202,636.80 16-31 January 65.52 290.36 84,916,593.00 376,311,420.00 1-15 February 57.00 163.90 73,868,630.40 212,414,486.40 16-28 February 63.86 183.13 82,764,952.62 237,341,763.69 1-15 March 59.86 125.88 77,582,348.31 163,140,579.69 16-31 March 61.24 69.36 79,369,921.11 89,884,777.85 1-15 April 76.75 92.04 99,466,006.15 119,285,036.31 16-30 April 82.19 103.91 106,522,825.85 134,673,341.54 1-15 May 75.24 76.28 97,515,725.54 98,864,363.08 16-31 May 76.92 81.40 99,690,831.00 105,488,730.00 1-15 June 84.01 87.67 108,870,480.00 113,623,171.20 16-30 June 66.82 67.24 86,599,411.20 87,141,657.60 1-15 July 75.61 79.59 97,985,289.60 103,154,083.20 16-31 July 71.80 78.22 93,055,149.00 101,372,472.00 1-15 August 53.20 57.56 68,953,248.00 74,595,686.40 16-31 August 54.46 54.49 70,581,196.80 70,625,347.20 1-15 September 40.16 64.67 52,045,027.20 83,809,641.60 16-30 September 48.15 91.61 62,396,092.80 118,732,694.40 The water of Walahar dam is distributed to irrigation network in Karawang district. This area belongs to Division 2 of PJT-II Jatiluhur. Based on the data 48 mentioned above, available water supply is higher than water utilized January to September 2006. North Tarum Networks West Tarum Networks East Tarum Networks Figure 16. Irrigation Network Map Source: PJT II Jatiluhur 49

4.1.14. Drought Indicator

Rain is the main source of water in the land for lives of crop, animal, and human. Decrease of land water can be identified by condition of surface cover and open surface. Based on evaporation fraction as indicator of drought level it could be concluded that actual rainfall data for Karawang district differed from time to time. For dry months the accumulation of rainfall is 100 mm or less per month. Actual occurrence of rainfall in several locations in Karawang is shown in Figure 17. Monthly average data from 1971 until 2000 are used. Monthly average rainfall data for 30 years indicate that the dry season in Karawang district occurs from May to September and rainy season from October to April. Rainfall in Batujaya 50 100 150 200 250 300 350 400 450 500 550 1 2 3 4 5 6 7 8 9 10 11 12 Month R a in fa ll m m Rainfall in Karaw ang 50 100 150 200 250 300 350 400 1 2 3 4 5 6 7 8 9 10 11 12 Month R a in fa ll m m 50 Rainfall in Jatisari 50 100 150 200 250 300 1 2 3 4 5 6 7 8 9 10 11 12 Month Ra in fa ll m m Figure 17. Monthly average rainfall pattern in Karawang district period from 1971 to 2000 Source: BMG According to the rainfall pattern the Landsat data used in this research can be grouped based on season as shown in Table 13. Table 13. Acquisition data base on condition of season Date Season 3-Jun-2003 Dry Season 18-Apr-2004 Wet Season 5-Jun-2004 Dry Season 23-Jul-2004 Dry Season 21-Apr-2005 Wet Season 10-Jul-2005 Dry Season 11-Aug-2005 Dry Season 1-Oct-2006 Dry Season Based on average rainfall condition the irrigation capacity decreases as shown, the decrease in ability of soil to save water, the leaves vegetation become yellowish adjusting to the long dry months. The decrease of irrigated water is correlated with increase of wide areas of drought vulnerability in Karawang district. Nevertheless, irrigation water supply from PJT II Jatiluhur especially in division II Karawang provides enough water and decreases drought potential areas. Result analysis for drought vulnerability in Karawang district by using satellite image Landsat ETM 7+ indicates that in the dry month only a small area is vulnerable of drought. Relatively wider areas of drought potential is detected at the beginning of rain season i.e. acquisition Landsat image in October 1, 2006. 51 Evaporative fraction EF values used to determine area of drought vulnerability range from 0 – 0.4. Landsat data June 3, 2003 analysis indicates potential of drought vulnerability is 2,746 Ha. Water availability in soil for June is adequate. Indicator of water availability can be seen by land cover condition with relatively green vegetation index and high energy used for evaporation that indicates water for evapotranspiration from soil and crop is available in the soil, only a small area in Karawang is potential of drought, which means most energy is used for sensible heat flux. The same result is shown from Landsat image analysis for 2004 and 2005. In 2004 there is increase of drought vulnerability from 1,074 Ha in June to 2,965 Ha in July. The same trend is observed for analysis in 2005 where increased area of drought is from 322 Ha in July to 1,229 Ha in August. Similar increase of drought vulnerability area is shown for result analysis in 2006. The analysis is based on data acquisition at the start of the rainy season, but climate conditions i.e. delay of the start of rainy season i.e. October is still the dry season. About 77,057 Ha areas in Karawang indicated drought vulnerability. The analysis results cover all areas of Karawang i.e. irrigated and non irrigated lands. Data of paddy filed areas obtained from the Department of Agriculture for Karawang district and after being overlaid between paddy fields as irrigated land and drought vulnerability area showed that the total area indicating drought vulnerability for irrigated land in June 2003, June 2004, July 2004, July 2005, August 2005 and October 2006 are 640 ha, 524 ha, 572 ha, 21 ha, 126 ha and 41,893 ha, respectively. 52 Figure 18. Paddy field areas in Karawang District Source: Department of Agriculture Based on statistical data as shown in Table 14 it indicates that area of drought vulnerability which could not be recovered by irrigation water from 2003 until 2006 has increased from 915 Ha in 2003 to 1,142 Ha in 2006. 53 Table 14. Drought vulnerability data in Karawang district Year Criteria 2003 2004 2005 2006 Prediction Drought Vulnerability in Ha - - 3,895 6,512 Drought Vulnerability in Ha 915 975 - 1,142 - : no data Details of predicted area of drought vulnerability in Karawang district are shown in Table 15. Table 15. Prediction of drought vulnerability Year No Sub District 2005 2006 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Tempuran Cileber Telagasari Lemah Abang Cilamaya Kulon Cilamaya Wetan Pedes Cibuaya Pakisjaya Batujaya Tirtajaya Pangkalan Rawamerta Jayakerta 345 235 115 250 260 1,350 648 380 175 45 100 1,274 625 460 1,030 470 1,350 200 785 318 Total 3,895 6,512 Source: PJT II Jatiluhur and Department Agriculture Karawang district Areas of drought vulnerability detected from analysis result of satellite data and predicted by the Department of Agriculture and PJT II Jatiluhur are different. The analysis result from satellite data in 2005 covered areas of Cikampek 63 ha, Jatisari 15 ha, and Klari 40 ha. In 2006, analysis result of satellite data of drought vulnerability areas covers all areas that are predicted drought vulnerability; however satellite detects wider areas of drought vulnerability. Referring to planting season and irrigation from PJT II Jatiluhur, in September all irrigated areas in Karawang already harvest their crop; so many paddy field areas are bare and intentionally let dry waiting for the next planting season. In October it is the start of the rainy season, but on October 1, 2006 no 54 rain is detected so there is still drought vulnerability. This matter is caused by planting pattern that is not based on group irrigation. Details of planting season, prediction of harvest and irrigation is presented in Table 16. Table 16. Schedule for planting season, prediction of harvest and watering irrigation. 55 According to statistical data in 2006, the total paddy field areas in Karawang is 94,385 ha, of which 91,063 ha is irrigated and 3,322 ha non irrigated lands. Details of paddy field areas from 2003 to 2006 are shown in Table 17. Irrigation system in Karawang leads to significant decrease of vulnerability drought areas. Water irrigation becomes an important factor in decreasing drought vulnerability areas in Karawang. Tabel 17. Paddy field area in Karawang district. Paddy Field 2003 2004 2005 2006 Irrigation 89,630 89,614 87,952 91,063 Non Irrigation 3,167 3,172 3,163 3,322 Total 92,797 92,786 91,115 94,385 Source : BPS Karawang Based on debit data and volume of water in 2006 in Walahar dam the water supply is higher than the water needed. Walahar dam itself regulates amount of water supplied to paddy field in irrigated lands in Karawang. Water flow rate ranges from 50-300 m 3 s, but consumption is about 40-90 m 3 s. Furthermore, water required for industrial and household PDAM in Karawang district is relatively low i.e. is 2.11 m 3 s and for agriculture during planting season is about 50-75 m 3 s. In September 2006 irrigation water used is relatively low and most of paddy field areas become dry till the next planting season. This condition causes areas of drought vulnerability detected become wider, not because of water deficit. Generally irrigation water is enough for agriculture during planting during the rainy rendeng and dry seasons gadu. Time for planting and harvesting in Karawang is different depending on paddy field classification based on irrigation. There are five paddy field classifications based on watering, planting and harvest seasons. Watering starts from group I and ends in group V. Almost all paddy field areas in Karawang depend on irrigation water. Group I is paddy field areas located near main irrigation channel, while group V is farther located with this pattern it is expected that all paddy field area need of water irrigation can be fulfilled. 56 In fact several paddy field areas in Karawang are still potential for drought vulnerability. Generally, areas suffer from drought vulnerability are located farthest from main irrigation channel and it is included in group V for watering. Based on field information there are three main causes of drought i.e.: 1. Water flows especially during dry season gadu cannot reach the paddy field area located farthest from main irrigation channel. 2. Irregular planting pattern that is in appropriate with watering classification of paddy field. 3. Irrigation infrastructure damage. Principally irrigation water is aimed to reduce discrepancy water supply during rainy and dry seasons. Inefficient use of water especially by farmers at the upper course near main irrigation channel causes difficulties for paddy fields located at the lower course to obtain water. The motivation factors are disobedience to planting time of appropriate classification of paddy field; inefficient water use that is allocated for other paddy field classification and infrastructure damage. Therefore, government involvement to prevent problems by repairing infrastructure and improving farmer knowledge about the importance of water distribution should be done. It is expected to minimize paddy field areas in Karawang district suffering from drought vulnerability as Karawang is one of the rice barns in Indonesia.

4.2. Web Development

4.2.1. Database Management

Database management software used for this system is MySQL 4.0.18 and this software has high capacity of storing data based on the operation system Window OS-32 bit and capacity is 4 GB. The logical design is a file that defines the basic organization of database. A data dictionary is a computer based catalog or dictionary containing metadata, which is data about data. A data dictionary includes a software component to manage database of data definition that is metadata about the structure. A data dictionary contains a list of all files in the database, the number of records in each file, and the names and types of each field. Most database management systems keep the data dictionary hidden from 57 users to prevent them from accidentally destroying its contents. Data dictionary do not contain any actual data from the database only the administrator could be authorized to manage it. Physical data model takes view of the data that describes how the data are to be physically stored and accessed on the storage devices of a computer system. For web-based GIS Drought Indicator system, phpMyAdmin 2.6.0 and MySQL 4.0.18 are used to develop a physical database design. Figure 19. Physical database using phpMyAdmin environment.

4.2.2. Code Constructing

Web page is created by PHP coding programming language. PHP Script can be written by using Macromedia Dreamweaver MX 2004 software. PHP is scripting language which is integrated into HTML page and it can run in server side. All the PHP syntax coded will be fully run in the server.

4.2.3. Web View

One of the main objectives of the thesis is to provide an easy to use interface that can be operated by a non-technical user. The interface is mainly menu driven, making it very easy to use. All the menu options are given appropriate, so that a novice computer user can use the system Figure 20. The demonstration of the basic features is available at http:adi.harry.or.id 58 M a i n M e n u Figure 20. Main interface of drought indicator at Karawang district From main menu the user can very easily select all information on theory about drought indicator, map from Web GIS, and other tools. Most of the commonly used menu option are provided as tools on a separate tool bar and can be used by clicking on the appropriate tool in the main menu. When the user clicks on a particular tool, new information is displayed in the same window.

4.2.4. Spatial Information from WEB GIS

Web based GIS Drought Indicator website was created by utilizing PHP script and map script. Figure 21 shows the screen shot of main page of web-based GIS drought indicator at Karawang district. Figure 21. Main page of web-based GIS drought indicator 59 This main page is equipped with a map to indicate the location of each level of drought indication. Map can be released to the Internet due to the integration of PHP script and map server. Web-based GIS system provided toolbar to make interactive information on spatial data such as zoom in, zoom out, full extent, re- center and query point. Spatial query option allows the user to retrieve attribute data from related database tables by selecting a location on the vector image displayed on the web browser. Interactive zoom query capabilities allow the user to view the displayed maps in greater details or to choose different areas of display. The other capability of web-based GIS is to select legend. In this system, information default shows a vector map of administration and raster image of drought indicator. So users can turn onoff those layers by a check list as the needed layer, then click Redraw Map, the new layers will be shown on the map area.

4.2.5. Data Editing

This system provided editing of all information on theory and all databases related to web information. Only the administrator is allowed to input, delete, and edit the data on the database. In this page the administrator is allowed to manage data and information about web and database. So the administrator can store news information and manage the comments. Figure 22. Editing page to database in the web system 60

4.2.6. Other Capability

The other capability is fully equipped application comment in the forum page and new information in the news page. On the forum page, users are allowed give to comments by using this system. Figure 23 shows the screen shot of add comment page of web system. Figure 23. The screen shot of add comment page of web system. The user must fill all text boxes provided by the administrator. The administrator put all announcements, notification or other information on the news page, but the users can only access brief or detailed information. Figure 24 shows the screen shot of add news page of web system. Figure 24. The screen shot of add news page of web system. 61 V. CONCLUSION AND RECOMENDATION

5.1 Conclusion

The drought potential from 2003 to 2005 at Karawang district is low either in the wet season or dry season. For rainy season data is taken in April and for dry season from June to October. The dominant drought indicated is 0.5-0.8, especially in paddy field areas. Based on drought indicators at Karawang district water is supplied by rainfall and irrigation available during dry or rainy season. In October 2006, the drought level is moderate until very dry as indicated by evaporative fraction which is about 0-0.4. In September 2006, all irrigated areas in Karawang already harvested their crop so many paddy field areas are bare and intentionally let dry waiting for the next planting session. October is usually the start of the rainy season, but in October 2006 there was no rainfall observed. Hence, wider drought vulnerability was detected. This research studied the implementation of a web-base GIS drought indicator at Karawang district using MapServer, available software, which was extended using MySQL database, PHP Script and MapScript to provide data and map services. This system has provided some good experience for understanding of the system application and gives a confidence to implement the knowledge and skill derived from learning process.

5.2 Recommendation