3/28/2010 Aplikasi SIG & Penginderaan Jauh  Penutupan lahan  Evaluasi lahan: -Land suitability : PENERAPAN MULTI CRITERIA DECISION MAKING (MCDM) DAN.

Presentasi berjudul: "3/28/2010 Aplikasi SIG & Penginderaan Jauh  Penutupan lahan  Evaluasi lahan: -Land suitability : PENERAPAN MULTI CRITERIA DECISION MAKING (MCDM) DAN."— Transcript presentasi:

1 3/28/2010 Aplikasi SIG & Penginderaan Jauh  Penutupan lahan  Evaluasi lahan: -Land suitability : PENERAPAN MULTI CRITERIA DECISION MAKING (MCDM) DAN GEOGRAPHICAL INFORMATION SYSTEM (GIS) PADA EVALUASI PERUNTUKAN LAHAN (Studi Kasus: DAS Ciliwung Hulu, Kab. Bogor, Jawa Barat) Dr. Syartinilia Wijaya Departemen Arsitektur Lanskap, FAPERTA-IPB Klasifikasi Data Citra untuk Penutupan Lahan  Proses Sebelum Klasifikasi Koreksi Geometri Koreksi Radiometri Koreksi Topografi  Klasifikasi Visual Terbimbing Tidak Terbimbing Element Order 2: Geometric Arrangement  Size – untuk menentukan obyek berdasarkan ukuran  Sifatnya adalah relatif  Shape – untuk membantu menentukan obyek berdasarkan bentuk  man made – cenderung garis lurus -Habitat Suitability: Landscape-scale habitat distributions of the Javan Hawk-Eagle (Spizaetus bartelsi) in Java Island, Indonesia VISUAL Element Order 1  Tone : Variasi kedalaman warna obyek dari hitam ke putih yang dapat dibedakan  Colour : Warna obyek Elements Order 2 Spatial Arrangement  Texture – frekuensi perubahan dan susunan dari tone  Pengamatan visual kehalusan/kekasaran (smoothness or roughness)  Misal Air : biasanya halus, Alang-alang : medium texture, and Hutan alam dataran rendah: kasar  Selalu ada pengecualian  natural – cenderung tidak beaturan  Pattern - arrangement spasial dari objects  Linear untuk jalan, sungai dll 1

2 3/28/2010 Element order 3 Locational or Positional  Site – bagaimana obyek berada pada suatu tempat  aspect, topography, geology, soil, vegetation and cultural Elements Order 3 Interpreted  Height – menjelaskan detail dari obyek  Tinggi pohon/bangunan features  Shadow – mungkin membantu/mengganggu interpretasi  Association – obyek biasanya berasosiasi dengan obyek yang lain.  Sangat membantu dalam interpretasi mand made obyek Analisis Visual Foto Udara  Identifikasi dapat ditingkatkan dengan informasi bayangan Klasifikasi Image/Citra Dimana Mangrove ? Dimana Hutan dataran rendah ? Dimana Perkebunan ? Dimana Lahan terbuka ? Dimana Sungai ? Dimana Jalan ? Dimana Awan ? Dimana Bayangan awan ? Dimana Semak belukar ? REFLECTANCE CURVE  Tujuan Klasifikasi Image:  untuk mengidentifikasi dan mengelompokkan digital number (DN) dari image yang menggambarkan objek/tipe penutupan lahan yang ada di bumi.  Klasifikasi citra merupakan bagian paling penting dari analisis citra digital (Digital image analysis).  Dua metode utama dalam klasifikasi citra: 1. Klasifikasi Terbimbing (Supervised Classification) 2. Klasifikasi tidak Terbimbing (Unsupervised Classification) 2

3 3/28/2010 Klasifikasi Terbimbing  Mengelompokan Nilai DN Berdasarkan dengan arahan operator  Menentukan rule/aturan pengelompokan Menentukan training area Tahapan dalam Klasifikasi Terbimbing Strategi klasifikasi citra  Minimum distance classifies image data on a database file using a set of 256 possible class signature segments as specified by signature parameter. Each segment specified in signature, for example, stores signature data pertaining to a particular class. Only the mean vector in each class signature segment is used. Other data, such as standard deviations and covariance matrices, are ignored (though the maximum likelihood classifier uses this).  Maximum likelihood Classification is a statistical decision criterion to assist in the classification of overlapping signatures; pixels are assigned to the class of highest probability. The maximum likelihood classifier is considered to give more accurate 3  Klasifikasi  Pengelompokan Ulang  Uji akurasi Training Area Klasifikasi Terbimbing Minimum Distance To Means Classification Stategy

4 4 3/28/2010 Probability Density Functions Defined by Maximum Likelihood Classifier  The parallelepiped classifier uses the class limits and stored in each class signature to determine if a given pixel falls within the class or not. The class limits specify the dimensions (in standard deviation units) of each side of a parallelepiped surrounding the mean of the class in feature space.  If the pixel falls inside the parallelepiped, it is assigned to the class. However, if the pixel falls within more than one class, it is put in the overlap class (code 255). If the pixel does not fall inside any class, it is assigned to the null class (code 0). Parallelepiped Classification Strategy Klasifikasi Tidak Terbimbing  Merupakan metode yang mengelompokkan piksel- piksel menjadi kelas penutupan lahan berdasarkan pengelompokan yang terjadi secara natural dari nilai image yang ada.  tidak ada campur tangan operator atau tanpa training area.  Tetap perlu pemahaman lapang untuk reklasifikasi Klasifikasi Tidak Terbimbing 1. 2. 3. 4. APLIKASI SIG What is at ? (Pertanyaan Locational) Where is it ? Where can I find Eagle ? Where is the suitable habitat ? (Pertanyaan Locational) How has it changed ? How is the impact of ? (Pertanyaan Conditional) What if vegetation change ? What if I develop new road ? (Pertanyaan Conditional)

5 ###### 3/28/2010 5 PENERAPAN MULTI CRITERIA DECISION MAKING (MCDM) DAN Lahan yang harus dilindungi dan non- lindung DAS Ciliwung Hulu, Kabupaten Bogor Kawasan budidaya Kawasan lindung Pertanian lahan basah Pertanian lahan kering Perkebunan teh Pemukiman Berdasarkan kriteria teknis dalam RTRW Bopunjur Optimalisasi Penggunaan lahan saat ini Konflik area ###### ######### # # #### # ### ## # ### # # # # # # # ## # # # # # GEOGRAPHICAL INFORMATION SYSTEM (GIS) PADA EVALUASI PERUNTUKAN LAHAN (Studi Kasus: DAS Ciliwung Hulu, Kab. Bogor, Jawa Barat) KEPPRES RI NO. 114 Tahun1999 Tentang Penataan Ruang Kawasan Bogor - Puncak-Cianjur # ## # # # # # # # # # Gr oun d tr uth poin t # GP S tra ck 2 002Su nga i C illiw ung Lege nda : N 0123 Kilometers Pulau Jawa Provinsi Jawa Barat Rekomendasi peruntukan lahan Kriteria Kawasan Lindung Sumber : - Ketentuan teknis kawasan lindung dalam RTRW Bopunjur (Bappeda, 2000) dengan berpedoman pada Keppres No.32 tahun 1990 tentang Pengelolaan Kawasan Lindung dan SK Gubernur Jawa Barat No. 413.21/SK.222-HUK/91 tentang Kriteria Lokasi dan Standar Teknis Penataan Ruang di Kawasan Puncak. Kesesuaian lahan untuk kawasan lindung Slope Daerah gerakan tanah tinggi ElevasiTanah Sempadan sungai Daerah tangkapan air utama >2000 m >40% Litosol, slope > 15% 100 m di kiri- kanan sungai Regosol, slope > 15% DAS Ciliwung Hulu, Kabupaten Bogor, Jawa Barat Kriteria Evaluasi Kesesuaian Lahan untuk Kawasan Budidaya Peruntukan Lahan Kriteria Pertanian lahan basah Slope (%) <3 Elevasi(m) 300-700 Drainase Terhambat Land cover Hutan Sedang-baik Cepat 3-8 9-15 16-25 700-1000 1000-1500 1500-2000 Perkebunan teh Semak belukar Ladang >25>2000Sawah Pemukiman Pertanian lahan kering <3300-700Baik-sedangHutan Cepat Terhambat 3-8 9-15 16-25 >25 700-1000 1000-1500 1500-2000 >2000 Perkebunan teh Semak belukar Ladang Sawah Pemukiman Perkebunan 1000-1500 1500-2000 >2000 Baik Sedang-cepat Terhambat <8 8-15 16-25 >25 Hutan Perkebunan teh Semak belukar Ladang Sawah Pemukiman <10 10-15 16-20 >20 300-700 700-1000 1000-1500 1500-2000 Baik Sedang Cepat Terhambat Pemukiman Hutan Perkebunan teh Semak belukar Ladang >2000Sawah Pemukiman Database Spasial Digital Elevation Model (DEM) Slope Aspect Drainase Penggunaan/penutupan Lahan

6 3/28/2010 6 Dalam aplikasi CP digunakan analisis sensitivitas untuk 3 nilai p yang berbeda yaitu p = 1, 2 dan(e.g. p>10). I p p i x ik d p 1 / p xi*xi* i 1 dp = kumpulan distance metrics = bobot yang ditetapkan untuk kriteria berdasarkan tingkat > 0, p = distance parameter, range dari 1 sampai i 1 i preferensi dimana x = tititk ideal p Metode Evaluasi Kesesuaian Lahan untuk Kawasan Budidaya Multi Criteria Decision Making (MCDM) Compromise Programming (CP) Perhitungan Nilai Distance Metric dalam Kerangka SIG  Nilai dan Bobot Struktur AHP Kawasan Budidaya Level 3: Nilai Kesesuaian lahan untuk pertanian lahan basah Slope (%) <3 3-8 >25 9-15 16-25 Drainase Terhambat Sedang-baik Cepat Elevasi (m) 300-700 700-1000 1000-1500 1500-2000 >2000 Penggunaan/pen utupan lahan Hutan Perkebunan teh Semak belukar Sawah Ladang Pemukiman Level 2: Bobot Teknik Perbandingan Berpasangan (Pairwise Comparison) Intensitas pentingnya 1 Definisi Equal: Kedua elemen yang dibandingkan sama pentingnya 3Moderate: Elemen yang satu sedikit lebih penting dibanding elemen yang lain 5Strong : Elemen yang satu sangat penting dibanding elemen lainnya 7979 Very strong: Satu elemen jelas lebih penting daripada elemen lainnya Extreme: Satu elemen mutlak lebih penting dibanding elemen lainnya 2,4,6,8 Kebalikan (½, 1 / 3, ¼..dst) Nilai-nilai antara diantara dua pertimbangan yang berdekatan Jika untuk elemen i mendapat satu angka bila dibandingkan dengan elemen j, maka j mempunyai nilai kebalikannya dibandingkan i Penghitungan Nilai dan Bobot Distribusi Kawasan Lindung dan Non-Lindung Model Kesesuaian Lahan Kawasan Budidaya Pertanian Lahan BasahPertanian Lahan Kering

7 Pemukiman 3/28/2010 “ Spizaetus bartelsi ” Perkebunan General IntroductionIntroduction Endemic raptor The Javan Hawk-Eagle (Spizaetus bartelsi) “Endangered status” “ Javan Hawk-Eagle ” “ Elang Jawa ” The Objectives The objective of this study is to extrapolate the predicted However, to be effective in the long term, JHE habitat management should not be confined only to the local and regional scale, but must also address the landscape-scale, i.e. the scale at which population processes occur. Since it is not feasible to conduct complete field surveys for a landscape-scale RS /GIS technologies, spatial analysis, modeling approaches Such models offer the possibility of being able to minimize field work and GIS- based models are easily updated as new information becomes available. Inductive Approach Southern part of West Java Model Extrapolation Landscape-scale: Java Island 7 Preliminary predictive habitat models were developed using logistic and autologistic regression (Syartinilia and Tsuyuki, 2008) Model Creation Local-scale: TNGP Model Validation Regional-scale: Study probability model of JHE habitat distribution from the local-scale model to landscape-scale model in order to generate map of potential and present habitat suitability for JHE in the entire landscape. Subsequently, population number of JHE will be estimated. West Java Regional-scale Landscape-scale: Java Island Local-scale Methods Area

8 ELEVATION (SLP) SLOPE (NDVI) 3/28/2010 8 Nest-site in Cugenang (CG) Male of JHE when perching on the tree in Gunung Baud Photo source: field observation, 2005-2006 Juvenile of JHE in front and back position New nest Old nest Nest-site in Gunung Baud (GB) Nest tree species: Quercus teysmannii Groundtruth check Nest tree species: Quercus teysmannii  GIS data sets  GeoCover Landsat mosaic, S-48-05_2000, S-49-05_2000.  Moderate Resolution Imaging Spectrometer (MODIS) NDVI 16-day, 250m of 2002 (MOD13Q1, Tile 28 & 29)  Shuttle Radar Topography Mission (SRTM) Digital terrain elevation data, 90m, 2000 (E105-E114; S5-S9)  Digital map of protected areas boundary in Java Data  Bird data sets  Bird Life International (2001); modified using some references and field survey. The best predicted probability model of JHE distribution Threshold probability value at 0.5 Habitat suitability model Model Accuracy Identification of habitat patches Population estimation Flowchart of this study 6.8592 1 0.0189 NDVI 19.8063 AUTOCOV Historical localities record after 1980 Probability value ≥ 0.5 Area > 20 km 2 Dividing the area of presumed suitable habitat by assumed home-range size Pi 1 exp 0.2371 SLP 0.0070 ELV Source: Syartinilia & Tsuyuki, 2008 AUTOCOVARIATEDifferent Vegetation (AUTOCOV) Normalize by 6*6 window size Index (ELV) 6.85921 exp 1 0.0189 NDVI 19.8063 AUTOCOV 0.2371 SLP 0.0070 ELV Pi The best predicted probability model Autologistic regression model using 1,500 m neighborhood size : The best predicted probability model of JHE distribution Model Accuracy Identification of habitat patches Population estimation Flowchart of this study 6.85921 exp 1 0.0189 NDVI 19.8063 AUTOCOV 0.2371 SLP 0.0070 ELV Pi Historical localities record after 1980 Probability value ≥ 0.5 Area > 20 km 2 Dividing the area of presumed suitable habitat by assumed home-range size Source: Syartinilia & Tsuyuki, 2008 Threshold probability value at 0.5 Habitat suitability model Predicted probability of JHE habitat distribution Results

9 of presumed suitable habitat by assumed home range size. Maximum home range size = 2,000 ha  several researchers had been used 3/28/2010 9 Totally 3,107 km² of suitable habitat of JHE in Java Island 41 locations (85%) of 49 historical localities record correctly predicted 7 locations (15%) omission error similar threats Suitable habitat of JHE based on threshold at 0.5 Suitable habitat Administrative boundary Historical localities record after 1980 Omission error location Distribution of habitat patches of JHE in Java Island Proportion area of habitat patches which located inside and outside protected area network: 60.4% existed inside the protected area network Population estimation Patch number LocationProvinceArea (km 2 ) Edge (km) Estimated population (pair) Minimum Maximum homerange 7 8 9 10 11 12 13 14 15 16 Mts Dieng (Mt. Sumbing) Mts Dieng (Mt. Sindoro) Mts Merapi-Merbabu Mt. Lawu Mt. Arjuno-Welirang Mt. Liman-Wilis Mt. Kawi Yang highlands Mts. Bromo Tengger Semeru National Park Mt. Raung Central Java East Java 54 55 127 212 117 81 336 401 123 62 55 54 165 312 193 89 666 577 168 14 32 53 29 20 84 100 31 3 6 11 6 4 17 20 6 Sum21663185542108 135Mean Median 325 Mt. of breeding West 95 128 24 5 Mts. Dieng (Mt. size West Java & 4 218 42 The 1 number Gede-Pangrango pairs of JHE has Java been estimated by dividing the area 2 Mt. Cireme West Java 56 73 14 3 3 Mt. Simpang-Mt.Tilu West Java 75 180 19 4 Minimum home range Kemulan) = 400 ha  Syartinilia 167 Tsuyuki (2008) 8 5 Mt. Papandayan West Java 100 108 25 5 6 Mt. Slamet Central Java 112 137 28 6 Discussions Habitat suitability model of JHE in Java island obtained in this study may be the most useful in the conservation planning process to help identify “hot spots” that are most likely to harbor JHE 60.4% of habitat patches were existed inside the protected areas and the remaining area extends far outside protected areas. The role of this area for the expansion of the JHE population is likely important. This evidence should be paid more attention from the several agencies (i.e. ministries, national parks, NGOs) to formulate a JHE conservation plan and to identify urgent conservation actions. Estimated population Source ALR_50 model (this study) Gjershaug et al. (2004) van Balen et al. (1999, 2000, 2001) Sözer and Nijman (1995) van Balen and Meyburg (1994) Meyburg et al. (1989) Pairs 108-542 270-600 137-188 81-108 67-81 60 Median 325 435 - The apparent discrepancy between this estimated population and others, which might not suggest as increase in present JHE increased accessibility to formerly unexplored habitat and more recent satellite imageries and GIS techniques application used in estimation of suitable habitat of JHE. The landscape-scale models based on extrapolation of the nest-site scale model using GIS/RS-based data could provide spatial explicit assessment of the potential and present habitat suitability at the scales of the greatest practical needs