Laporkan Masalah

Multiple Endmember Spectral Mixture Analysis (MESMA) untuk Pemetaan Spesies Invasif Acacia decurrens di Taman Nasional Gunung Merbabu Menggunakan Citra PRISMA Hyperspectral

Anggita Sulistyarini, Dr. Sanjiwana Arjasakusuma, S.Si, M.GIS.

2023 | Skripsi | KARTOGRAFI DAN PENGINDRAAN JAUH

Salah satu spesies vegetasi invasif yang mendominasi ekosistem hutan hujan tropika pegunungan rendah dan pegunungan tinggi di Taman Nasional Gunung Merbabu (TNGMb) adalah Acacia decurrens. Spesies invasif dapat mendominasi suatu area secara cepat sehingga menyebabkan penurunan populasi jenis asli serta mengganggu stabilitas ekosistem. Oleh karena itu, pengendalian persebaran spesies invasif diperlukan agar biodiversitas dapat terjaga. Pengendalian spesies invasif menggunakan penginderaan jauh hiperspektral dinilai lebih efisien daripada metode konvensional karena mampu menyajikan cakupan spasial yang luas, multi waktu, dan multi saluran. Multiple Endmember Spectral Mixture Analysis (MESMA) diperlukan dalam proses spectral unmixing karena sebaran komunitas Acacia decurrens berukuran lebih kecil daripada resolusi spasial citra PRISMA Hyperspectral. Model yang dibangun dalam penelitian ini menggunakan berbagai variasi data spectral library dan citra PRISMA Hyperspectral. Hasil uji akurasi kelas umum menunjukkan bahwa model spectrometer and smoothed image data variasi a dan variasi b merupakan model terbaik dengan nilai overall accuracy sebesar 85,48%. Hasil model terbaik menunjukkan bahwa spesies invasif Acacia decurrens banyak ditemukan pada sisi di hampir semua arah mata angin wilayah kajian pada ketinggian sekitar 1800 – 2400 mdpl. Dominasi Acacia decurrens paling banyak berada pada sisi barat daya, barat, dan barat laut TNGMb. 

One of the invasive species that dominates the lowland and highland topical rainforest ecosystem in Taman Nasional Gunung Merbabu (TNGMb) is Acacia decurrens. Invasive species can rapidly take over an area, leading to a decline in native species populations and disrupting ecosystem stability. Consequently, controlling the spread of invasive species is imperative to preserve biodiversity. Hyperspectral remote sensing has been deemed more efficient compared to conventional methods as it can provide broad spatial coverage, multi-temporal data, and multi-bands information. Multiple Endmember Spectral Mixture Analysis (MESMA) plays a crucial role in the process of spectral unmixing due to the fact that the distribution of Acacia decurrens communities is smaller than the spatial resolution of PRISMA Hyperspectral Image. The model constructed in this study utilized various variations of spectral library data and PRISMA Hyperspectral Image. The outcomes of general accuracy testing revealed that the spectrometer and smoothed image data models with variations a and b performing the best models, achieving overall accuracy of 85.48%. The best models indicate that the invasive species Acacia decurrens is predominantly distributed on almost all sides of the study area at elevations ranging from 1800 to 2400 meters above sea level with the highest dominance of Acacia decurrens is on the southwestern, western, and northwestern sides of TNGMb.

Kata Kunci : PRISMA Hyperspectral, MESMA, spesies invasif Acacia decurrens

  1. S1-2023-441713-abstract.pdf  
  2. S1-2023-441713-bibliography.pdf  
  3. S1-2023-441713-tableofcontent.pdf  
  4. S1-2023-441713-title.pdf