Indonesia is the world’s fourth largest coffee producer. Coffee plantations cover 1.2 million ha of the country with a production of 500 kg/ha. However, information regarding the distribution of coffee plantations in Indonesia is limited. This study aimed to assess the accuracy of classification model and determine its important variables for mapping coffee plantations. The model obtained 29 variables which derived from the integration of multi-resolution, multi-temporal, and multi-sensor remote sensing data, namely, pan-sharpened GeoEye-1, multi-temporal Sentinel 2, and DEMNAS. Applying a random forest algorithm (tree = 1000, mtry = all variables, minimum node size: 6), this model achieved overall accuracy, kappa statistics, producer accuracy, and user accuracy of 79.333%, 0.774, 92.000%, and 90.790%, respectively. In addition, 12 most important variables achieved overall accuracy, kappa statistics, producer accuracy, and user accuracy 79.333%, 0.774, 91.333%, and 84.570%, respectively. Our results indicate that random forest algorithm is efficient in mapping coffee plantations in an agroforestry system.

In 1997, the worst forest fire in Indonesia occurred and hit mangrove forest areas including in Sembilang National Park Banyuasin Regency, South Sumatra. Therefore, the Indonesian government keeps in trying to rehabilitate the mangrove forest in Sembilang National Park. This study aimed to identify the mangrove forest changing and to predict on the future year. The situations before and after forest fire were analyzed. This study applied an integrated Markov Chain and Cellular Automata model to identify mangrove forest change in the interval years of 1989–2015 and predict it in 2028. Remote sensing technology is used based on Landsat satellite imagery (1989, 1998, 2002, and 2015). The results showed mangrove forest has decreased around 9.6% from 1989 to 1998 due to forest fire, and has increased by 8.4% between 1998 and 2002, and 2.3% in 2002–2015. Other results show that mangroves area has continued to increase from 2015 to 2028 by 27.4% to 31% (7974.8 ha). It shows that the mangrove ecosystem is periodically changing due to good management by the Indonesian government.

Abstrak Hutan mangrove Petengoran adalah ekowisata mangrove di Provinsi Lampung yang terancam punah karna meningkatnya aktivitas pengunjung. Sehingga, informasi persebaran mangrove sangat diperlukan untuk tujuan konservasi. Dewasa ini, banyak peneliti memanfaatkan teknologi pengindraan jauh untuk pemetaan mangrove menggunakan algoritma machine learning. Penelitian ini bertujuan untuk membandingkan algoritma support vector machine (SVM) dan random forest (RF) untuk pemetaan mangrove menggunakan komposit RGB dan NDVI pada citra Sentinel 2A. Hasil penelitian menunjukkan bahwa algoritma RF memberikan akurasi yang lebih tinggi dibandingkan dengan SVM dibuktikan dengan nilai akurasi keseluruhan dan indeks kappa RF sebesar 92,68% dan 0,88, sedangkan pada SVM sebesar 91,86% dan 0,87. Meski demikian, terdapat kesalahan klasifikasi hutan mangrove di kedua algoritma. Hal tersebut disebabkan oleh kemiripan spektral jenis tanaman dan tidak adanya efek topografi. Sehingga, penelitian selanjutnya diharapkan dapat menambahkan efek topografi untuk mendapatkan akurasi yang lebih baik. Abstract The Petengoran is mangrove ecotourism in Lampung Province which is threatened with extinction due to increased community activity. Thus, information on the distribution of the Petengoran mangroves is needed for conservation. Many researchers use remote sensing technology to map mangroves using machine learning algorithms. This study aims to compare SVM and random forest RF algorithms for mapping mangroves using RGB and NDVI composites on Sentinel 2A imagery. The results showed that the RF algorithm provides higher accuracy compared to SVM. This is evidenced by the overall accuracy value and RF kappa index of 92.68% and 0.88, while the SVM is 91.86% and 0.87. However, there is a misclassification of mangrove forests in both algorithms. This is due to the spectral similarity of vegetation and no topographical effect. Thus, future research is expected to add topographical effects to obtain higher accuracy.

Oil palm plantations are the largest oil-producing plant and are included in the multipurpose plant category as a contributing economy and reducing the poverty. Oil palm age is one of the factors that support the growth and production process of oil palm. We used Landsat 8 Operational Land Imager (OLI) to estimate the oil palm age. The aims in this study are to analyse the relationship between NDVI (Normalized Difference Vegetation Index) and oil palm age in variation of age by using logarithmic regression and to estimate the distribution of oil palm age. Methodology in this study are collecting of Landsat 8, preprocessing, making of Regions of Interest for each of oil palm age, making of relationship analysis between NDVI and oil palm age, and estimating the oil palm age. The study shows that NDVI have positive correlation which R² = 0,66 and equation y = 0.0425 ln(x) + 0.723 means the higher NDVI value the higher the age. The result of this study show that the majority of age in the study area is 0-5 year.

Indonesia adalah negara kepulauan terbesar dengan populasi yang tersebar di berbagai wilayah, memiliki kondisi geografis dan karakteristik masyarakatnya yang beragam. Hal ini menjadi peluang dalam melaksanakan pemerataan pembangunan di negera ini khususnya di tingkat desa dalam menjalankan program pemberdayaan. Namun permasalahannya adalah banyak desa yang belum memiliki peta sebagai dasar perencanaan pembangunan desa, khususnya desa Candimas. Sehingga tujuan pengabdian ini adalah pembuatan peta garis Desa Candimas. Metode yang digunakan adalah studi literatur, sosialisasi, pengumpulan data, digitasi, dan penyajian peta. Hasil dari pengabdian ini berupa peta garis Desa Candimas skala 1:7500 yang menginformasikan penggunaan lahan diantaranya adalah sawah, fasilitas umum, bangunan, vegetasi, pemukiman, dan perkebunan. Peta tersebut dimanfaatkan sebagai bahan identifikasi wilayah-wilayah yang memerlukan perhatian khusus, sehingga program pemberdayaan dapat dijalankan dengan lebih efektif dan efisien

Coffee plantation is one of the main agricultural sectors in the world, especially in Indonesia. For sustainable coffee management, it is important to obtain an assessment of the spatial distribution of coffee plantation. In fact, most of the coffee trees planted under the forest canopy as jungle coffee. However, remote sensing studies devoted to coffee in Indonesia have been limited due to spectral similarity with forest. Hence, the shaded coffee plantation has been difficult to be mapped. This condition takes aerial photography as an alternative solution to detect the distribution of shaded coffee plantation with high spatial resolution. This paper presents an assessment of the classification method of aerial photography for detecting shaded coffee plantation. The data used in this study is aerial photography captured in a shaded coffee field located in Gunung Puntang. Some classification methods namely Pixel-Based and Object-Based Image Analysis (OBIA) were used. Those method performances were evaluated by comparing the classification result as overall accuracy. The result shows that OBIA is the best method for detecting shaded coffee plantation. It produces an overall accuracy of 73.01%.