<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" version="2.0">
<channel>
<title>UF - Marine Science And Technology</title>
<link>http://repository.ipb.ac.id/handle/123456789/110</link>
<description/>
<pubDate>Thu, 19 Nov 2026 16:27:23 GMT</pubDate>
<dc:date>2026-11-19T16:27:23Z</dc:date>
<item>
<title>Identifikasi Daun Mangrove dan Lamun secara Otomatis melalui Deep Learning dan Pengembangan Aplikasi Berbasis Website</title>
<link>http://repository.ipb.ac.id/handle/123456789/180134</link>
<description>Identifikasi Daun Mangrove dan Lamun secara Otomatis melalui Deep Learning dan Pengembangan Aplikasi Berbasis Website
Sari, Sabrina Tijani
Ekosistem mangrove dan lamun memainkan peran penting dalam keseimbangan pesisir, sehingga memerlukan metode identifikasi dan pemantauan yang efektif. Identifikasi spesies secara tradisional memakan waktu, membutuhkan banyak tenaga kerja, dan bergantung pada keahlian pengamat. Penelitian ini mengembangkan model deep learning menggunakan algoritma YOLOv11 untuk mengidentifikasi empat spesies vegetasi pesisir (Avicennia alba, Rhizophora apiculata, Enhalus acoroides, dan Thalassia hemprichii) secara otomatis dan mengimplementasikannya ke dalam aplikasi berbasis web. Dataset sebanyak 1.500 gambar berlabel bounding-box dibagi menjadi data training, validasi, dan testing (80:10:10). Dievaluasi melalui confusion matrix, model YOLOv11 mencapai nilai mAP50 sebesar 0,994, dengan presisi berkisar antara 0,97-0,99 dan recall dari 0,92-1,00 pada seluruh kelas. Model diintegrasikan ke dalam situs website berbasis Flask, di mana fungsi-fungsi utamanya berjalan sesuai rancangan. Penelitian ini menyoroti potensi AI dan aplikasi website untuk identifikasi vegetasi pesisir secara otomatis serta penyediaan data dalam kerangka kerja OHI.; Mangrove and seagrass ecosystems play a crucial role in coastal balance, requiring effective identification and monitoring methods. Traditional species identification is time-consuming, labor-intensive, and reliant on observer expertise. This study develops a deep learning model using the YOLOv11 algorithm to automatically identify four coastal vegetation species (Avicennia alba, Rhizophora apiculata, Enhalus acoroides, and Thalassia hemprichii) and implements it into a web-based application. A dataset of 1.500 bounding-box labeled images was split into training, validation, and test sets (80:10:10). Evaluated via a confusion matrix, the YOLOv11 model achieved a mAP50 of 0,994, with precision ranging from 0,97 to 0,99 and recall from 0,92 to 1,00 across classes. The model was integrated into a Flask-based website, where core functions performed as designed. This research highlights the potential of AI and web applications for automated coastal vegetation identification and data provision within the OHI framework.
</description>
<pubDate>Thu, 01 Jan 2026 00:00:00 GMT</pubDate>
<guid isPermaLink="false">http://repository.ipb.ac.id/handle/123456789/180134</guid>
<dc:date>2026-01-01T00:00:00Z</dc:date>
</item>
<item>
<title>Hubungan Antara Kandungan Nitrat dan Fosfat Dengan Kelimpahan Fitoplankton di Perairan Pulau Harapan, Kepulauan Seribu a.n RIFTIA ISHAB ANNISA (C5401221069)</title>
<link>http://repository.ipb.ac.id/handle/123456789/180041</link>
<description>Hubungan Antara Kandungan Nitrat dan Fosfat Dengan Kelimpahan Fitoplankton di Perairan Pulau Harapan, Kepulauan Seribu a.n RIFTIA ISHAB ANNISA (C5401221069)
Annisa, Riftia Ishab
Abstrak: Perairan Pulau Harapan, Kepulauan Seribu, menghadapi tekanan&#13;
antropogenik yang berpotensi meningkatkan konsentrasi unsur hara dan memicu&#13;
eutrofikasi. Penelitian ini bertujuan menganalisis kandungan nitrat dan fosfat,&#13;
mengidentifikasi jenis dan kelimpahan fitoplankton, serta mengkaji hubungan&#13;
keduanya. Pengambilan data dilakukan pada Februari, Mei, Agustus, dan&#13;
November 2025 di empat Stasiun. Nitrat dan fosfat dianalisis secara&#13;
spektrofotometri, sedangkan fitoplankton diidentifikasi dan dihitung&#13;
menggunakan Sedgwick–Rafter Counting Chamber. Konsentrasi nitrat berkisar&#13;
0,018–0,070 mg/L dan fosfat 0,010–0,023 mg/L. Nitrat melampaui baku mutu&#13;
selama seluruh periode pengamatan, sedangkan fosfat mulai Mei. Fitoplankton&#13;
didominasi Bacillariophyceae, dengan 7–18 taksa dan kelimpahan 1,0–9,8 × 106&#13;
sel/L. Fosfat menunjukkan hubungan lebih kuat terhadap kelimpahan fitoplankton&#13;
(R² = 0,1555) dibandingkan nitrat (R² = 0,0079). Rasio Redfield menunjukkan nitrat&#13;
sebagai unsur hara pembatas.&#13;
&#13;
&#13;
Abstract: Pulau Harapan, Kepulauan Seribu, is subject to anthropogenic pressures that&#13;
may increase nutrient concentrations. This study aimed to determine nitrate and&#13;
phosphate concentrations, identify phytoplankton composition and abundance, and&#13;
their relationships. Data were collected in February, May, August, and November&#13;
2025 at four stations. Nitrate and phosphate were analyzed spectrophotometrically,&#13;
while phytoplankton were enumerated using a Sedgwick–Rafter Counting&#13;
Chamber. Nitrate ranged from 0.018–0.070 mg/L and phosphate from 0.010–0.023&#13;
mg/L. Nitrate exceeded the seawater standard throughout observations, whereas&#13;
phosphate exceeded the standard from May onward. Phytoplankton were&#13;
dominated by Bacillariophyceae, with 7–18 taxa and abundances of 1.0–9.8 × 106&#13;
cells/L. PCA explained 51.68% of variance. Regression showed a stronger&#13;
relationship between phosphate and abundance (R² = 0.1555) than nitrate (R² =&#13;
0.0079). The N:P ratio remained below 16, indicating nitrate limitation.
</description>
<pubDate>Thu, 01 Jan 2026 00:00:00 GMT</pubDate>
<guid isPermaLink="false">http://repository.ipb.ac.id/handle/123456789/180041</guid>
<dc:date>2026-01-01T00:00:00Z</dc:date>
</item>
<item>
<title>Deteksi Invertebrata Laut pada Ekosistem Lamun menggunakan eDNA Metabarcoding di Pantai Sanur, Bali</title>
<link>http://repository.ipb.ac.id/handle/123456789/180034</link>
<description>Deteksi Invertebrata Laut pada Ekosistem Lamun menggunakan eDNA Metabarcoding di Pantai Sanur, Bali
Alfian, Muhammad Rizky
Ekosistem lamun Pantai Sanur, Bali mengalami tekanan antropogenik akibat aktivitas wisata dan limbah domestik yang berpotensi memengaruhi komunitas invertebrata laut. Penelitian ini bertujuan mengidentifikasi keanekaragaman spesies menggunakan metode environmental DNA (eDNA) metabarcoding dengan penanda gen 16S rRNA. Sampel air dianalisis melalui ekstraksi DNA, PCR (Polymerase Chain Reaction), Next-Generation Sequencing (NGS), serta analisis QIIME 2 dan BLAST (Basic Local Alignment Search Tool). Hasil penelitian mengidentifikasi 16 spesies dari 21 Operational Taxonomic Units (OTUs) yang mewakili 9 kelas dan 7 filum. Nilai Indeks Shannon-Wiener sebesar 2,03–2,15 menunjukkan keanekaragaman sedang. Metode eDNA metabarcoding terbukti efektif untuk mendeteksi keanekaragaman invertebrata laut pada ekosistem lamun.; The seagrass ecosystem at Sanur Beach, Bali, is subject to anthropogenic pressure from tourism activities and domestic waste, which may affect marine invertebrate communities. This study aimed to identify species diversity using the environmental DNA (eDNA) metabarcoding method with the 16S rRNA gene marker. Water samples were analyzed through DNA extraction, PCR (Polymerase Chain Reaction), Next-Generation Sequencing (NGS), and analysis using QIIME 2 and BLAST (Basic Local Alignment Search Tool). The results identified 16 species from 21 Operational Taxonomic Units (OTUs), representing 9 classes and 7 phyla. Shannon-Wiener Index values ??ranging from 2.03 to 2.15 indicated moderate diversity. The eDNA metabarcoding method proved effective for detecting marine invertebrate diversity in the seagrass ecosystem.
</description>
<pubDate>Thu, 01 Jan 2026 00:00:00 GMT</pubDate>
<guid isPermaLink="false">http://repository.ipb.ac.id/handle/123456789/180034</guid>
<dc:date>2026-01-01T00:00:00Z</dc:date>
</item>
<item>
<title>Model Sebaran Tumpahan Minyak di Lintas Utara Selat Malaka</title>
<link>http://repository.ipb.ac.id/handle/123456789/179982</link>
<description>Model Sebaran Tumpahan Minyak di Lintas Utara Selat Malaka
Ramadhana, Naurah Alifah
Selat Malaka bagian utara merupakan jalur pelayaran padat dan rentan terhadap risiko tumpahan minyak, namun studi pemodelan trayektori tumpahannya masih terbatas. Penelitian ini bertujuan memodelkan sebaran dan nasib tumpahan minyak hipotetik di lintas utara Selat Malaka menggunakan General NOAA Operational Modeling Environment (GNOME), dengan data arus dan pasang surut dari Copernicus Marine Service (CMEMS) serta data angin dari ERA5 sebagai gaya penggerak, yang divalidasi terhadap data observasi arus, pasut, dan angin primer. Simulasi dilakukan pada dua jenis minyak, Arabian Light dan Swallow Ace, mewakili musim utara dan musim selatan pada empat kondisi fase pasut. Hasil validasi menunjukkan keandalan tinggi pada data arus dan pasut, sedangkan data angin memiliki keterbatasan akibat resolusi data observasi. Pola arus dan angin musiman memengaruhi luas dan arah sebaran tumpahan, dengan musim selatan menghasilkan sebaran lebih luas namun musim utara menghasilkan fraksi terdampar lebih tinggi. Nasib minyak Arabian Light didominasi oleh evaporasi, sedangkan Swallow Ace didominasi oleh fraksi terapung akibat sifat persisten dan viskositas tinggi. Analisis tumpang susun terhadap ekosistem pesisir menunjukkan risiko lebih tinggi terhadap terumbu karang dibandingkan hutan mangrove, dengan risiko meningkat pada musim selatan.; The northern Strait of Malacca is a heavily trafficked shipping lane vulnerable to oil spill risk, yet trajectory modeling studies for this region remain limited. This study aims to model the dispersion and fate of a hypothetical oil spill in the northern Strait of Malacca using the General NOAA Operational Modeling Environment (GNOME), by current and tidal data from the Copernicus Marine Service (CMEMS) and wind data from ERA5, validated against primary current, tidal, and wind observations. Simulations were conducted for two oil types, Arabian Light and Swallow Ace, representing the northeast and southwest monsoon seasons under four tidal phase conditions. Validation results showed high reliability for current and tidal data, while wind data showed limitations due to the resolution of observational data. Seasonal current and wind patterns influenced the extent and direction of the spill, with the southwest monsoon producing wider dispersion while the northeast monsoon produced a higher stranded fraction. Arabian Light's fate was dominated by evaporation, whereas Swallow Ace was dominated by the floating fraction due to its persistent and highly viscous nature. Overlay analysis with coastal ecosystems indicated higher risk to coral reefs than to mangrove forests, with risk increasing during the southwest monsoon.
</description>
<pubDate>Thu, 01 Jan 2026 00:00:00 GMT</pubDate>
<guid isPermaLink="false">http://repository.ipb.ac.id/handle/123456789/179982</guid>
<dc:date>2026-01-01T00:00:00Z</dc:date>
</item>
</channel>
</rss>
