ESTIMASI POPULASI OWA JAWA (HYLOBATES MOLOCH) MELALUI PENDEKATAN SIDIK JARI VOKAL DI PUSAT PENDIDIKAN KONSERVASI ALAM BODOGOL, JAWA BARAT

ZAHRA AFRIANI AFIF SUSANTO, . (2026) ESTIMASI POPULASI OWA JAWA (HYLOBATES MOLOCH) MELALUI PENDEKATAN SIDIK JARI VOKAL DI PUSAT PENDIDIKAN KONSERVASI ALAM BODOGOL, JAWA BARAT. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text COVER.pdf Download (3MB)
	Text BAB 1.pdf Download (205kB)
	Text BAB 2.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text BAB 3.pdf Restricted to Registered users only Download (752kB) | Request a copy
	Text BAB 4.pdf Restricted to Registered users only Download (2MB) | Request a copy
	Text BAB 5.pdf Restricted to Registered users only Download (193kB) | Request a copy
	Text DAFTAR PUSTAKA.pdf Download (227kB)
	Text LAMPIRAN.pdf Restricted to Registered users only Download (2MB) | Request a copy

Abstract

Kelestarian owa jawa (Hylobates moloch) terancam oleh fragmentasi habitat yang memicu ketimpangan distribusi dan fluktuasi kepadatan di sepanjang sisa tutupan hutan. Perolehan data populasi menjadi tantangan dalam upaya pelestarian, mengingat sifatnya yang intoleran terhadap kehadiran manusia. Status terancam punah pada IUCN Red List dan pencantuman dalam Apendiks I CITES memperkuat urgensi pengembangan metode perhitungan populasi yang efisien. Penelitian ini bertujuan untuk menjabarkan pola vokalisasi sebagai dasar identifikasi individu dan mengestimasi populasi secara non-invasive melalui pendekatan bioakustik dengan teknik passive acoustic monitoring (PAM) di Bodogol, Taman Nasional Gunung Gede Pangrango. Data diambil pada Januari–Februari 2026 menggunakan enam unit AudioMoth v.1.2.0. Analisis melalui Raven Pro v.1.6 menerapkan ambang batas signal-to-noise ratio (SNR) ≥25 dB dan band-pass filter 400–1.600 Hz. Identifikasi dilakukan melalui ekstraksi mel-frequency cepstral coefficients (MFCCs), klasterisasi affinity propagation (AP) dan visualisasi uniform manifold approximation and projection (UMAP). Hasil menunjukkan dominasi song bout betina sebesar 98,8%. Vokalisasi betina membentuk struktur great call meliputi fase pre-trill, trill, dan termination. Fase pre-trill terbukti paling informatif dalam mengklasterisasi individu. Algoritma AP menghasilkan 18 klaster pada plot UMAP yang merepresentasikan 18 individu betina dewasa dari kelompok berbeda. Berdasarkan rata-rata ukuran kelompok, total estimasi populasi di Bodogol adalah 59 individu. Rata-rata frekuensi deteksi mencapai 37,2% dengan titik tertinggi di Stasiun Penelitian Bodogol sebesar 60%. Temuan ini membuktikan bahwa sidik jari vokal yang berlandaskan pada karakteristik vokalisasi betina berpotensi sebagai dasar identifikasi individu. Pendekatan PAM yang dikombinasikan dengan klasterisasi MFCCs efektif sebagai metode estimasi populasi non-invasive untuk mendukung monitoring jangka panjang dan strategi konservasi owa jawa. ***** The conservation of the Javan gibbon (Hylobates moloch) is now threatened by habitat fragmentation, which leads to uneven distribution and fluctuations in population density across the remaining forest cover. Collecting population data poses a challenge for conservation efforts, given the species’ intolerance of human presence. Its Endangered status on the IUCN Red List and listing in CITES Appendix I underscore the urgency of developing efficient population estimation methods. This study aims to describe vocalization patterns as a basis for individual identification and to estimate population size non-invasively through a bioacoustic approach using passive acoustic monitoring (PAM) techniques in Bodogol, Gunung Gede Pangrango National Park. Data were collected in January–February 2026 using six AudioMoth v.1.2.0 units. Analysis using Raven Pro v.1.6 applied a signal-to-noise ratio (SNR) threshold of ≥25 dB and a band-pass filter of 400–1,600 Hz. Identification was performed using mel-frequency cepstral coefficients (MFCCs), affinity propagation (AP) clustering, and uniform manifold approximation and projection (UMAP) visualization. The results showed a 98.8% dominance of female song bouts. Female vocalizations formed a great call structure comprising pre-trill, trill, and termination phases. The pre-trill phase proved to be the most informative for clustering individuals. The AP algorithm generated 18 clusters on the UMAP plot, representing 18 adult females from different groups. Based on the average group size, the total estimated population in Bodogol is 59 individuals. The average detection frequency reached 37.2%, with a peak of 60% at Bodogol Research Center. These findings demonstrate that vocal fingerprints based on female vocalization characteristics have the potential to serve as a basis for individual identification. The PAM approach, combined with MFCC clustering, is effective as a non-invasive population estimation method to support long-term monitoring and conservation strategies for the Javan gibbon.

Item Type:	Thesis (Sarjana)
Additional Information:	1). Mohamad Isnin Noer, M.Si; 2). Prof. Dr. Ratna Komala, M.Si.
Subjects:	Sains > Matematika > Software, Sistem Informasi Komputer Sains > Ilmu Bumi > Biologi Sains > Zoologi Sains > Mikro Biologi Sains > Mikro Biologi > Ekologi
Divisions:	FMIPA > S1 Biologi
Depositing User:	Users 36471 not found.
Date Deposited:	09 Sep 2026 01:39
Last Modified:	09 Sep 2026 01:39
URI:	http://repository.unj.ac.id/id/eprint/72883

Actions (login required)

View Item