Assessment of land use change in the Thuma forest reserve region of Malawi, Africa

A study was conducted in Thuma area in central Malawi to quantify contemporary land cover and to explore the degree of land use change in the Thuma forest reserve area of Malawi by analysing and comparing satellite-derived land cover maps from 1997, 2007 and 2017. The study was carried out using Remote Sensing and Geographic Information System (GIS), focusing on analysis of Landsat 5 ETM and Landsat 8 ORI/TIRS satellite images. The classification was conducted for the following distinct classes; closed forest, open forest, shrubland, savanna grassland, agriculture fields, and water. The analysis revealed that closed forest diminished from 19% in 1997 to 10% in 2007 to 6% in 2017. Open forest reduced from 30% to 21% from 1997 to 2007 but increased to 22% in 2017. Agriculture area almost doubled from 37 % in 1997 to 64 % in 2017. Actual area from 1997 to 2017, shows that closed forest has reduced from 7,000 ha to 3,000 ha while open forest from 12,900 ha to 7800 ha. Savanna grassland has doubled from 5,900 ha to 13,000 ha. However, future studies should use modern satellites such as Sentinel and Landsat 9 for improved quantification of changes. The findings show that even the protected forest reserve (previously dominated by closed forest) is not fully protected from deforestation by local communities. Government and other stakeholders should devise measures to meet the needs of the surrounding communities and the ecological/biophysical needs of the reserves. Based on this study, issues of re-demarcation of the forest reserve and accessed area should also be explored. This study serves as a reference for the management of Thuma Forest Reserve as a refuge for natural tree species, rivers that harbour endemic fish species (Opsaridium microlepis and Opsaridium microcephalis) and the sustainable management of endangered elephants in the reserve.

Automatic extraction of land cover statistics from satellite imagery by deep learning1

In this paper, we address the challenge of producing fully automated land cover estimates from satellite imagery through Deep Learning algorithms. We developed our system according to a tile-based, classify-and-count design. To implement the classification engine of the system, we adopted a cutting-edge Convolutional Neural Network model named Inception-V3, which we customized and trained for scene classification on the EuroSAT dataset. We tested and validated our system on two Sentinel-2 images representing quite different Italian territories (with an area of 751 km2 and 443 km2, respectively). Because no genuine ground-truth is available for the land cover of these sub-regional territories, we built a pseudo ground-truth by integrating land cover information from flagship European projects CORINE and LUCAS. A critical and careful analysis shows that our automatic land cover estimates are in good agreement with the pseudo ground-truth and offers extensive evidence of the remarkable segmentation ability of our system. The limits of our approach are also critically discussed in the paper and possible countermeasures are illustrated. When compared with traditional projects like CORINE and LUCAS, our automatic land cover estimation system exhibits three fundamental advantages: it can dramatically reduce production costs; it can allow delivering very timely and frequent land cover statistics; it can enable land cover estimation for very small territorial areas, well beyond the NUTS-2 level. As an additional outcome of land cover estimation, our system also automatically generates moderate resolution land cover maps that might be used in cartography projects as an agile first-level tool for map update or change detection purposes.

Developing Multi-Source Indices to Discriminate between Native Tropical Forests, Oil Palm and Rubber Plantations in Indonesia

Over the last 18 years, Indonesia has experienced significant deforestation due to the expansion of oil palm and rubber plantations. Accurate land cover maps are essential for policymakers to track and manage land change to support sustainable forest management and investment decisions. An automatic digital processing (ADP) method is currently used to develop land cover change maps for Indonesia, based on optical imaging (Landsat). Such maps produce only forest and non-forest classes, and often oil palm and rubber plantations are misclassified as native forests. To improve accuracy of these land cover maps, this study developed oil palm and rubber plantation discrimination indices using the integration of Landsat-8 and synthetic aperture radar Sentinel-1 images. Sentinel-1 VH and VV difference (>7.5 dB) and VH backscatter intensity were used to discriminate oil palm plantations. A combination of Landsat-8 NDVI, NDMI with Sentinel-1 VV and VH were used to discriminate rubber plantations. The improved map produced four land cover classes: native forest, oil palm plantation, rubber plantation, and non-forest. High-resolution SPOT 6/7 imagery and ground truth data were used for validation of the new classified maps. The map had an overall accuracy of 92%; producer’s accuracy for all classes was higher than 90%, except for rubber (65%), and user’s accuracy was over 80% for all classes. These results demonstrate that indices developed from a combination of optical and radar images can improve our ability to discriminate between native forest and oil palm and rubber plantations in the tropics. The new mapping method will help to support Indonesia’s national forest monitoring system and inform monitoring of plantation expansion.

Land Cover Changes Based on Cellular Automata for Land Surface Temperature in Semarang Regency

Land cover changes based on cellular automata for surface temperature in Semarang Regency has increased significantly due to the continuous rise in its population. Therefore, this study aims to identify, analyze and predict multitemporal land cover changes and surface temperature distribution in 2028. Data on the land cover map were obtained from Landsat 7 and 8 based on supervised classification, while Land Surface Temperature (LST) was calculated from its thermal bands. The collected data were analyzed for accuracy through observation, while Cellular Automata - Markov Chain was used to predict the associated changes in 2028. The result showed that there are 4 land cover maps with 5-year intervals from 2003 to 2018 at an accuracy of more than 85%. Furthermore, the existing land covers were dominated by forest with decreasing trend, while the built-up area continuously increased. The existing Land surface temperature range from 20.6°C to 36.6°C, at an average of 28.2°C and a yearly increase of 0.07°C. The temperature changes are positively correlated with the occurrence of land conversion. Land cover predictions for 2028 show similar forest dominance, with a 23,4% built-up area at a surface temperature of 28.9°C. Keywords: Land cover change; Cellular Automata-Markov Chain; Land Surface Temperature Copyright (c) 2021 Geosfera Indonesia and Department of Geography Education, University of Jember This work is licensed under a Creative Commons Attribution-Share A like 4.0 International License

Identifikasi dan Pencegahan Daerah Rawan Bencana Kebakaran Hutan dan Lahan Gambut Di Kalimantan Barat

Kalimantan Barat memiliki potensi bencana kebakaran hutan dan lahan gambut yang tinggi karena banyaknya titik api dan jenis lahan gambut yang mudah terbakar pada musim kemarau. Tujuan dari penelitian ini adalah untuk memetakan dan menentukan kecenderungan titik pamas dan mengidentifikasi dan mencegah kawasan rawan kebakaran hutan dan lahan gambut dengan data hotspot, peta curah hujan, peta tutupan lahan, peta kesatuan hidrologis gambut, dan peta cekungan air tanah menggunakan Sistem Informasi Geografis atau SIG. Metode overlap digunakan untuk menganalisis kecenderungan titik panas sedangkan Overlay dan Scoring digunakan untuk mengidentifikasi kawasan rawan kebakaran hutan dan lahan. Setelah dilakukan analisis titik panas, terdapat kecenderungan curah hujan pada kelas curah hujan 1.500-3.000 mm/tahun dengan 2.192 kejadian. Perubahan tutupan lahan di kawasan hutan mengalami penurunan sebesar 7,96%. Peningkatan tutupan lahan di kawasan non-hutan sebesar 11,26%, mempengaruhi potensi dan kecenderungan titik api dan bencana kebakaran hutan dan lahan. Kubu Raya memiliki tingkat kerawanan bencana kebakaran pada kelas sangat rawan dengan luasan 0,26%, dan Kapuas Hulu memiliki tingkat kerawanan bencana kebakaran pada kelas tidak rawan dengan luas 0,19%. Kabupaten Ketapang merupakan daerah dengan tingkat pencegahan tertinggi, dengan luas cekungan airtanah sebesar 26,46%.ABSTRACTWest Kalimantan has a high potential for forest and peatland fire disasters due to the high number of hotspots and the type of peatland which burns easily during the dry season. The purpose of this research is to map and determine the trend of hotspots and areas prone to forest and peatland fires and prevent them with hotspot data, rainfall maps, land cover maps, maps of peat hydrological units, and maps of groundwater basins using Geographic Information Systems or GIS. The overlap method is used to analyze the trend of hotspots; meanwhile, Overlay and Scoring are used to identify areas prone to forest and land fires in this research. After analyzing the hotspots, there is a tendency for rainfall with a class of 1,500-3,000mm/year with 2,192 events. Land cover change in forested areas decreased by 7.96%. It increased land cover in non-forest areas by 11.26%, affecting the potential and tendency of hotspots and forest and land fire disasters. Kubu Raya has a fire disaster vulnerability level in the very vulnerable class with an area of 0.26%, and Kapuas Hulu has a fire disaster vulnerability level in the non-prone class with an area of 0.19%. Ketapang Regency is the area with the highest prevention rate, with a groundwater basin area of 26.46%.

A Sensitivity Assessment of COSMO-CLM to Different Land Cover Schemes in Convection-Permitting Climate Simulations over Europe

The question of how sensitive the regional and local climates are to different land cover maps and fractions is important, as land cover affects the atmospheric circulation via its influence on heat, moisture, and momentum transfer, as well as the chemical composition of the atmosphere. In this study, we used three independent land cover data sets, GlobCover 2009, GLC2000 and ESACCI-LC, as the lower boundary of the regional climate model COSMO-CLM (Consortium for Small Scale Modeling in Climate Mode, v5.0-clm15) to perform convection-permitting regional climate simulations over the large part of Europe covering the years 1999 and 2000 at a 0.0275° horizontal resolution. We studied how the sensitivity of the impacts on regional and local climates is represented by different land cover maps and fractions, especially between warm (summer) and cold (winter) seasons. We show that the simulated regional climate is sensitive to different land cover maps and fractions. The simulated temperature and observational data are generally in good agreement, though with differences between the seasons. In comparison to winter, the summer simulations are more heterogeneous across the study region. The largest deviation is found for the alpine area (−3 to +3 °C), which might be among different reasons due to different classification systems in land cover maps and orographical aspects in the COSMO-CLM model. The leaf area index and plant cover also showed different responses based on various land cover types, especially over the area with high vegetation coverage. While relating the differences of land cover fractions and the COSMO-CLM simulation results (the leaf area index, and plant coverage) respectively, the differences in land cover fractions did not necessarily lead to corresponding bias in the simulation results. We finally provide a comparative analysis of how sensitive the simulation outputs (temperature, leaf area index, plant cover) are related to different land cover maps and fractions. The different regional representations of COSMO-CLM indicate that the soil moisture, atmospheric circulation, evaporative demand, elevation, and snow cover schemes need to be considered in the regional climate simulation with a high horizontal resolution.

Comparing floral resource maps and land cover maps to predict predators and aphid suppression on field bean

Context Predatory insects contribute to the natural control of agricultural pests, but also use plant pollen or nectar as supplementary food resources. Resource maps have been proposed as an alternative to land cover maps for prediction of beneficial insects. Objectives We aimed at predicting the abundance of crop pest predating insects and the pest control service they provide with both, detailed flower resource maps and land cover maps. Methods We selected 19 landscapes of 500 m radius and mapped them with both approaches. In the centres of the landscapes, aphid predators – hoverflies (Diptera: Syrphidae), ladybeetles (Coleoptera: Coccinellidae) and lacewings (Neuroptera: Chrysopidae) – were surveyed in experimentally established faba bean phytometers (Vicia faba L. Var. Sutton Dwarf) and their control of introduced black bean aphids (Aphis fabae Scop.) was recorded. Results Landscapes with higher proportions of forest edge as derived from land cover maps supported higher abundance of aphid predators, and high densities of aphid predators reduced aphid infestation on faba bean. Floral resource maps did not significantly predict predator abundance or aphid control services. Conclusions Land cover maps allowed to relate landscape composition with predator abundance, showing positive effects of forest edges. Floral resource maps may have failed to better predict predators because other resources such as overwintering sites or alternative prey potentially play a more important role than floral resources. More research is needed to further improve our understanding of resource requirements beyond floral resource estimations and our understanding of their role for aphid predators at the landscape scale.

A Dual Network for Super-Resolution and Semantic Segmentation of Sentinel-2 Imagery

There is a growing interest in the development of automated data processing workflows that provide reliable, high spatial resolution land cover maps. However, high-resolution remote sensing images are not always affordable. Taking into account the free availability of Sentinel-2 satellite data, in this work we propose a deep learning model to generate high-resolution segmentation maps from low-resolution inputs in a multi-task approach. Our proposal is a dual-network model with two branches: the Single Image Super-Resolution branch, that reconstructs a high-resolution version of the input image, and the Semantic Segmentation Super-Resolution branch, that predicts a high-resolution segmentation map with a scaling factor of 2. We performed several experiments to find the best architecture, training and testing on a subset of the S2GLC 2017 dataset. We based our model on the DeepLabV3+ architecture, enhancing the model and achieving an improvement of 5% on IoU and almost 10% on the recall score. Furthermore, our qualitative results demonstrate the effectiveness and usefulness of the proposed approach.

Land cover changes monitoring over ten years in upstream watershed of Deli Serdang Regency North Sumatra Province

Deli Serdang is one of the regencies in North Sumatra Province, experiencing relatively rapid development and population. Increasing in demand for the availability of land as living space. Two sub-districts of upstream watershed experienced changes in land cover, namely; Sinembah Tanjung Muda (STM) Hilir and STM Hulu. Monitoring changes in land cover in both sub-districts is essential, given that they are located in the upstream area of the watershed and will impact other areas in the lower watershed. This study aims to analyse land cover changes in both sub-districts over ten years (2009 - 2019). The method used in calculating land changes that occur is change detection. Field surveys were carried out to ensure that the land cover conditions on the land cover maps followed the field’s actual conditions. The research shows the period of 2009 – 2019, land cover that has increased in the area are mining, industry, open land, settlements, livestock and shrubs. The decrease in the area occurred in land cover, including dryland forest, mixed gardens and cultivated land.