Search

Land cover maps are important documents for local governments to perform urban planning and management. A field survey using measuring instruments can produce an accurate land cover map. However, this method is time-consuming, expensive, and labor-intensive. A number of researchers have proposed using remote sensing, which generates land cover maps using an optical satellite image with various statistical classification procedures. Recently, artificial intelligence (AI) technology, such as deep learning, has been used in multiple fields, including satellite image classification, with satisfactory results. In this study, a WorldView-2 image of Terangun in Aceh Province, which was acquired on Aug 2, 2016, was classified using a commonly used deep-learning-based classification, namely, U-net. There were eight classes used in the experiment: building, road, open land (such as green open space, bare land, grass, or low vegetation), river, farm, field, aquaculture pond, and garden. For comparison, three classification methods: maximum-likelihood, random forest, and support vector machine, were performed compared to U-Net. A land cover map provided by the government was used as a reference to evaluate the accuracy of land cover maps generated using two classification methods. The results with 100 randomly selected pixels revealed that U-Net was able to obtain a 72% and 0.585 for overall and kappa accuracy, respectively; whereas, overall accuracy and kappa accuracy for the maximum likelihood, random forest and support vector machine methods were 49% and 0.148; 59% and 0.392; and 67% and 0. 511; respectively. Therefore, U-Net outperformed those three of classification methods in classifying the image.

As the most populous island globally, Java Island has various vulnerabilities to disasters ranging from geological to hydro-meteorological. One of the most common hydro-meteorological disasters is the drought that occurs every year in the dry season. This disaster causes crop failure, land and forest fires, and clean water shortages. In this study, the Sea and Land Surface Temperature Radiometer (SLSTR) instrument onboard the Sentinel-3 platform was used to map drought using the Vegetation Temperature Condition Index (VTCI) algorithm based on the scattering space technique of Land Surface Temperature (LST) and Normalized Different Vegetation Index (NDVI). During 2018, the highest and lowest LST occurred on September 28 (309 oK), and February 1 (278 oK); the highest and lowest NDVI occurred on July 1 (0.466) and November 13 (0.221). In comparison, the driest conditions indicated the lowest VTCI (0.162) on July 17 and the wettest on August 28 (0.508). The driest and wettest situation co-occurred: at the end of the dry session and the rainy session, it shows that the rain greatly contributes to high VTCI. A regular drought mapping needs to be performed as an effort to disaster risk reduction. Drought maps are then used as a spatial recommendation in reforestation intervention to reduce drought in the future.

Koreksi atmosfer merupakan proses yang sangat penting karena efek atmosfer mampu mempengaruhi gelombang elektromagnetik dari matahari ke objek dan dari objek ke sensor yang menyebabkan terjadinya kesalahan pada data citra. Salah satu metode koreksi atmosfer yang sering digunakan adalah metode 6SV (Second Simulation of the Sensor Signal in the Solar Spectrum-Vector). Landsat-8 adalah citra satelit yang sering dimanfaatkan karena kemudahan perolehan data yang gratis. Pada tahun 2014, USGS mengeluarkan produk level tinggi untuk surface reflectance (Reflectance-BOA). Dengan keluarnya produk ini, maka kendala koreksi atmosfer yang harus dilakukan menjadi berkurang. Namun, belum diketahui seberapa besar tingkat akurasi dari produk ini. Pada penelitian ini akan dilakukan uji akurasi dari produk Landsat-8 surface reflectance menggunakan data in-situ Remote sensing reflectance (Rrs (λ)) dan Suspended Solid (SS) yang direkam secara langsung di 5 (lima) stasiun sepanjang Danau Kasumigaura, Jepang. Sebagai perbandingan, data tersebut akan dibandingkan dengan data Landsat-8 yang dikoreksi dari efek atmosfer menggunakan parameter koreksi dari hasil simulasi menggunakan metode 6SV. Dari hasil penelitian ini, menunjukkan bahwa nilai Rrs (λ)-6SV dan Rrs (λ)-L memiliki nilai yang lebih tinggi dari Rrs (λ)-insitu di sepuluh stasiun. Sedangkan untuk korelasinya (R2) Rrs (λ)-L memiliki korelasi lebih tinggi yaitu sebesar 0,926 dibandingkan dengan Rrs (λ)-6SV yang hanya 0,123. Besarnya NMAE serta RMSE yang dihasilkan Rrs (λ)-6SV dan Rrs (λ)-L ini secara berturut-turut adalah 140,262%; 0,012 dan 140,061%; 0,008.

Anoa (Bubalus sp.) is one of the endemic species of Sulawesi Island, which is protected because it is threatened with extinction. This study aims to determine the possible distribution of anoa habitat based on environmental parameters by considering the influence of human disturbance. Human disturbance to Anoa is in the form of settlement and transportation infrastructure. The purpose of this paper is to observe whether human disturbance has an effect or not on Anoa spatial distribution. Using 98 encounter species data points, the model discovered that settlement and transportation infrastructure contributed only 4.3% and 1.4%, respectively. Meanwhile, elevation, temperature, and water have a dominant contribution that exceeds 90%. This data indicated that the anoa habitat is less sensitive to human disturbance.

Drought monitoring is essential to detect the presence of drought, and the comprehensive change of drought conditions on a regional or global scale. This study used satellite precipitation data from the Tropical Rainfall Measuring Mission (TRMM), but refined the data for drought monitoring in Java, Indonesia. Firstly, drought analysis was conducted to establish the standardized precipitation index (SPI) of TRMM data for different durations. Time varying SPI spatial downscaling was conducted by selecting the environmental variables, normalized difference vegetation index (NDVI), and land surface temperature (LST) that were highly correlated with precipitation because meteorological drought was associated with vegetation and land drought. This study used time-dependent spatial regression to build the relation among original SPI, auxiliary variables, i.e., NDVI and LST. Results indicated that spatial downscaling was better than nonspatial downscaling (overall RMSEs: 0.25 and 0.46 in spatial and nonspatial downscaling). Spatial downscaling was more suitable for heterogeneous SPI, particularly in the transition time (R: 0.863 and 0.137 in June 2019 for spatial and nonspatial models). The fine resolution (1 km) SPI can be composed of the environmental data. The fine-resolution SPI captured a similar trend of the original SPI. Furthermore, the detailed SPI maps can be used to understand the spatio-temporal pattern of drought severity.

The satellite-based regression model provides the data model that identifies water quality for inland and coastal waters. However, the satellite regression usually depends on the selection of observation, satellite data, and model type. A resampling simulation technique, such as sequential simulation using geographically weighted regression (GWR simulation), can be applied in generating multiple realizations for water quality estimation to reduce the sampling effect and consider spatial heterogeneity. Traditional models often result in considerable underestimation in extreme observations. The GWR simulation provides the best goodness of fit and spatial varying relationship between observed water quality and remote sensing considering parameter outlier and noise removal for parameter stability. This simulation model can increase the sampling diversity from various observations and reduce the neighboring effects of observations using outlier and noise removal. The model that handles spatial uncertainty and heterogeneity is a novel tool for inferring the characteristics of water quality from a series of sample subsets.

Lake Matano and Towuti are two of 15 lake priorities in Indonesia. For preservation purposes, a routine water quality monitoring from satellite is needed. In this study, 11 scenes of Landsat 8 data were processed to produce chlorophyll-a concentration as an indicator of water quality condition on these two lakes. The result showed that water quality in Lake Matano and Towuti were in low cholorphyll-a condition with chlorophyll-a concentration ranged from 0.000– 2.298 mg/m3, 0.000-2.236 mg/m3, respectively. The accuracy of estimated Chlorophyll-a in these two lakes were affected by an inaccurate of atmospheric corrected data. To improve the accuracy, a more accurate atmospheric correction algorithm for Landsat 8 was still required.

Regional water quality mapping is the key practical issue in environmental monitoring. Global regression models transform measured spectral image data to water quality information without the consideration of spatially varying functions. However, it is extremely difficult to find a unified mapping algorithm in multiple reservoirs and lakes. The local model of water quality mapping can estimate water quality parameters effectively in multiple reservoirs using spatial regression. Experiments indicate that both models provide fine water quality mapping in low chlorophyll-a (Chla) concentration water (study area 1, root mean square error, RMSE: 0.435 and 0.413 mg m−3 in the best global and local models), whereas the local model provides better goodness-of-fit between the observed and derived Chla concentrations, especially in high-variance Chla concentration water (study area 2, RMSE: 20.75 and 6.49 mg m−3 in the best global and local models). In-situ water quality samples are collected and correlated with water surface reflectance derived from Sentinel-2 images. The blue-green band ratio and Maximum Chlorophyll Index (MCI)/Fluorescence Line Height (FLH) are feasible for estimating the Chla concentration in these waterbodies. Considering spatially-varying functions, the local model offers a robust approach for estimating the spatial patterns of Chla concentration in multiple reservoirs. The local model of water quality mapping can greatly improve the estimation accuracy in high-variance Chla concentration waters in multiple reservoirs.

Sebagai salah satu dengan arus transportasi pelayaran dan pengeboran sumur minyak dan gas yang ramai, Perairan Pantura Jawa sangat rentan terhadap peristiwa pencemaran laut akibat dari peristiwa tumpahan minyak. Karena itu diperlukan sebuah metode yang dapat mendeteksi persebaran tumpahan minyak dengan cepat guna meminimalisir dampak negatif terhadap lingkungan. Penginderaan jauh aktif sangat berguna dalam deteksi cepat ini karena memiliki kemampuan untuk melakukan akuisisi data secara independen dari kondisi cuaca dan kemampuan akuisisi pada siang maupun malam hari. Penelitian ini bertujuan untuk melakukan analisis deteksi dan monitoring persebaran tumpahan minyak menggunakan citra satelit Sentinel-1A dengan menerapkan tools oil spill detection pada software SNAP yang mengidentifikasi area hitam menggunakan nilai ambang batas threshold. Diperlukan data tambahan berupa arah dan kecepatan angin yang didapatkan dari metode Wind Field Estimation untuk membedakan area tumpahan minyak. Luas tumpahan minyak di perairan Pantura Jawa pada Bulan Juli, Agustus dan September 2019 sebesar 18,79 km², 15,08 km², dan 19,6804 km², sementara pada Bulan Januari 2020 sebesar 8,62 km².

Surabaya merupakan kota terbesar kedua di Indonesia setelah Jakarta, dimana jumlah penduduk kota Surabaya setiap tahunnya selalu mengalami kenaikan. Peningkatan populasi penduduk di wilayah perkotaan dapat mengubah pola ruang kawasan perkotaan. Pengunaan lahan akan bergeser dari keperluan pertanian menjadi keperluan tempat tinggal, kawasan bisini/industri dan aktivitas lainnya. Perubahan tutupan lahan ini akan berdampak pada kondisi iklim dan cuaca di kawasan perkotaan sehingga menyebabkan terjadinya fenomena Urban Heat Island (UHI). Pemantauan fenomena ini diperlukan di perkotaan yang mengalami perkembangan pesat, termasuk di Kota Surabaya. Pada penelitian ini akan dianalisis perubahan distribusi Urban Heat Island (UHI) di Kota Surabaya pada tahun 2002, 2014, dan 2019 menggunakan Citra Landsat 7 (ETM+) dan Citra Landsat 8 (OLI/TIRS). Langkah awal yang dilakukan berupa perhitungan Land Surface Temperature (LST) dengan metode Single-Channel Algorithm, kemudian dilakukan analisis distribusi UHI dengan metode Ambang Batas dan Hot Spot Analysis (Getis-Ord Gi*). Hasilnya didapatkan nilai suhu permukaan rata-rata Kota Surabaya untuk tahun 2002, 2014, dan 2019 secara berurutan sebesar 29,094°C; 26,889°C; dan 27,130°C. Uji korelasi Pearson Product Moment dilakukan antara LST dengan suhu lapangan, diperoleh hasi koefisien korelasi (Rxy) sebesar 0,449. Terakhir, dari peta distribusi UHI metode Ambang Batas, luas area yang terdampak UHI selalu mengalami penurunan dari tahun 2002 ke 2014, dan 2014 ke 2019 dengan luas penurunan masing-masing sebesar 0,760 km2 dan 7,995 km2. Hal yang sama juga terjadi untuk metode Hot Spot Analysis, dengan luas penurunan pada tahun 2002 ke 2014, dan 2014 ke 2019 sebesar 2,027 km2 dan 31,168 km2.

Fenomena urbanisasi yang cepat di zaman modern ini menimbulkan banyak masalah di berbagai aspek, salah satunya aspek lingkungan, khususnya di wilayah perkotaan seperti Kota Surabaya. Kota ini merupakan salah satu wilayah metropolitan yang menjadi pusat perekonomian dengan pertumbuhan yang cepat. Seiring dengan pertambahan populasi dan percepatan pembangunan, optimalisasi fungsi ruang terbuka hijau (RTH) sebagai penyerap emisi karbon menjadi hal penting untuk diperhatikan. Oleh sebab itu, penelitian ini ditujukan untuk mengetahui persebaran keanekaragaman hayati pada pohon dan potensinya dalam penyerapan emisi karbon di salah satu wilayah padat, yaitu Kelurahan Darmo, Kota Surabaya. Pengamatan spesies pohon dilakukan menggunakan data orthophoto dan light detection and ranging (lidar). Untuk mengetahui jumlah pohon berdasarkan tutupan kanopinya, penelitian ini menggunakan algoritma local maxima, serta metode klasifikasi berbasis objek untuk menentukan spesiesnya. Dengan memanfaatkan kedua data melalui metode tersebut, dihasilkan tingkat kesesuaian klasifikasi dengan akurasi kappa sebesar 0,64. Sejumlah 5.429 dari 6.778 pohon yang ada di Kelurahan Darmo teridentifikasi sebagai individu pohon yang diketahui spesiesnya, terdiri dari: Albizia julibrissin, Cerbera manghas, Ficus benjamina, Plumeria rubra, Polyalthia longifolia, Pterocarpus indicus, Roystonea regia, Tamarindus indica, dan Terminalia catappa. Albizia julibrissin merupakan spesies pohon yang mendominasi di kelurahan ini dengan indeks nilai penting (INP) sebesar 0,42 berdasarkan hasil analisis dominansi vegetasi. Melalui perhitungan average nearest neighbor, diketahui bahwa 80,0% kondisi setiap spesies pohon di kelurahan ini memiliki kecenderungan pola persebaran yang menggerombol. Berdasarkan jumlah dan spesies pohon yang diketahui, maka diduga Kelurahan Darmo memiliki cadangan karbon sejumlah 574,8 ton C. Pterocarpus indicus merupakan spesies pohon yang paling berpengaruh terhadap penyimpanan cadangan karbon di kelurahan tersebut dengan jumlah 96,7 ton C atau 16,8% dari keseluruhan cadangan karbon di kelurahan tersebut. Jumlah cadangan karbon di Kelurahan Darmo jika dikonversi menjadi kemampuan pohon dalam penyerapan emisi karbon dioksida setara dengan 2.107,7 ton CO2.

COVID-19 is spreading into Indonesia and has reached tens of thousands of cases as of September 30, 2020. It was recommended by the American Public Health Association and the Centers for Disease Control and Prevention to remain physically active during COVID-19 quarantine by regularly visiting parks and green spaces as it can protect the body against the consequences of quarantine impacting physical and mental health. In this research, green space was monitored by using remotely sensed data. The green space distribution was obtained from the calculation of the Greenness Index from Landsat-8 Surface Reflectance Tier 1 satellite imagery processed through the Google Earth Engine platform. This study was conducted to determine the value of Greenness Index (GI), Case Fatality Rate (CFR) value due to COVID-19, and the relationship between them in 42 sub-districts in DKI Jakarta in the period of April to September 2020. Twenty-eight subdistricts (67%) showed negative correlation values that indicated that more green space in a region affects lower CFR growth.

Java is Indonesia’s and the world’s most populous island. The increase in population on the island of Java reduces the area of forest and other vegetation covers. Landslides, floods, and other natural disasters are caused by reduced vegetation cover. Furthermore, it has the potential to lead to the extinction of flora and fauna. The Normalized Difference Vegetation Index (NDVI) can be used to monitor the vegetation cover. This study analyzes the NDVI changes value from 2005 to 2020 using Terra and Aqua MODIS image data processed using Google Earth Engine. Processing was carried out in some stages: down-setting, performing NDVI processing, calculating monthly average NDVI, calculating annual average NDVI, and analyzing. From the study results, the NDVI value of Terra and Aqua MODIS data has a solid but imperfect correlation coefficient due to differences in orbital time which causes differences in solar zenith angle, sensor viewing angle, and azimuth angle. Then from this study, it was found that overall, changes in vegetation density cover on the island of Java decreased, which was indicated by the NDVI decline rate of -0.00047/year. The most significant decrease in NDVI value occurred in the period 2015–2016, covering an area of 13994.630 km2, and the most significant increase in NDVI occurred in the period 2010–2011, covering an area of 2256.101 km2.

To investigate the long-term trend of light conditions in Lake Kasumigaura (a shallow eutrophic lake with high turbidity) in Japan, 215 images of MERIS data from the period 2003–2012 were processed at four stations using a semi-analytical algorithm to retrieve their inherent optical properties after atmospheric correction. Previously obtained Secchi depths (SDs) were somewhat underestimated by the proposed algorithms, and the ratio of the predicted SD to the measured SD changed with the ratio of tripton to chlorophyll a. A submodel was then built describing the ratio of scattering to backscattering based on the ratio of tripton to chlorophyll a (a trend supported by a number of previous reports) and applied to the prediction of SD in this lake. The model showed a gradually increasing trend at all stations in the predicted SD over the period, which was validated by the observations. The relationship between the measured and predicted SDs within a 2-day period was scattered, but showed a positive correlation at a significant level. In addition, this proposed method with the submodel describing the ratio of scattering to backscattering was applied to in situ reflectance spectra, and a correlation at a significant level was confirmed between the measured and predicted SDs.

The spatial heterogeneity and nonlinearity exhibited by bio-optical relationships in turbid inland waters complicate the retrieval of chlorophyll-a (Chl-a) concentration from multispectral satellite images. Most studies achieved satisfactory Chl-a estimation and focused solely on the spectral regions from near-infrared (NIR) to red spectral bands. However, the optical complexity of turbid waters may vary with locations and seasons, which renders the selection of spectral bands challenging. Accordingly, this study proposes an optimization process utilizing available spectral models to achieve optimal Chl-a retrieval. The method begins with the generation of a set of feature candidates, followed by candidate selection and optimization. Each candidate links to a Chl-a estimation model, including two-band, three-band, and normalized different chlorophyll index models. Moreover, a set of selected candidates using available spectral bands implies an optimal composition of estimation models, which results in an optimal Chl-a estimation. Remote sensing images and in situ Chl-a measurements in Lake Kasumigaura, Japan, are analyzed quantitatively and qualitatively to evaluate the proposed method. Results indicate that the model outperforms related Chl-a estimation models. The root-mean-squared errors of the Chl-a concentration obtained by the resulting model (OptiM-3) improve from 11.95 mg&middot, m-3 to 6.37 mg&middot, m-3, and the Pearson&rsquo, s correlation coefficients between the predicted and in situ Chl-a improve from 0.56 to 0.89.

Indonesian food security for chicken production as staple protein foods is essential for ensuring the availability to be consumed by toddlers, teenagers, and adults. Ensuring a good price to farmers and a stable of chicken food for the customers also should be achieved. Chicken (meat and eggs) food is most consumed than beef food in Indonesia. Hence, this paper develops a modeling of chicken production for Indonesian protein food security to capture the causality between variables in corn supply, feed mill plant, chicken production, and customer demand system. Modeling of chicken production using Geographic Information System (GIS) approach and developing of system dynamics model are used in this study. The findings of this study suggests that corn productivity, feed mill production, fodder consumption for chicken, and chicken meat and eggs demand and stock as well are critical variables within the model.

Penginderaan jauh memegang peran penting untuk mendukung berbagai sektor pembangunan dan pelestarian lingkungan. Pada pelaksanaannya penginderaan jauh tidak lepas dari penggunaan citra satelit untuk mendapatkan informasi dipermukaan bumi. Citra Landsat 8 bisa didapatkan dengan mengunduh melalui web earth explorer. Data yang didapat berupa data mentah. Data ini masih perlu dilakukan proses dulu untuk dapat digunakan. Selain itu data citra satelit merupakan data raster yang memiliki ukuran penyimpanan data ( storage ) yang besar. Terlebih lagi untuk melakukan pengolahan citra satelit dengan cepat membutuhkan spesifikasi komputer yang tinggi. Penelitian ini bertujuan melakukan koreksi radiometrik ( radiomtric correction ) citra Landsat 8, setelah itu membuat dashboard data penginderaan jauh dimana data yang disediakan berupa citra reflektan permukaan ( surface reflectance ), sehingga data yang diunduh dari dashboard siap digunakan. Metode koreksi radiometrik menggunakan metode Apparent Reflectance ArcGIS, sedangkan pembuatan dashboard menggunakan platform GeoNode. Hasil penelitian ini berupa website yang mampu menampilkan dan mengunduh citra yang sudah dikoreksi radiometrik. Website tersebut baru bisa diakses komputer host dengan membuka alamat IP komputer virtual melalui browser. Raster yang bisa ditampilkan GeoNode adalah raster integer , sehingga citra reflektan permukaan yang berupa raster float harus diubah dulu menjadi raster integer . Untuk mengubah citra reflektan permukaan menjadi raster integer digunakan faktor skala 10000.

An empirical approach through remote sensing generally produces a robust data model of water quality for inland and coastal water. Traditional regressions in water quality mapping fail because the bio-optical relationship of turbid water exhibits nonlinear and heterogeneous patterns. In addition, in situ data are generally insufficient in the water quality mapping. Mapping based on a relatively small amount of water quality samples is considered a practical issue in environmental monitoring. Learning-based algorithms that require a large amount of data are inapplicable in this case. According to the concept of Nadaraya–Watson estimator, the kernel model can estimate nonlinear and spatially varying water quality maps effectively in turbid water. Experiments indicate that the kernel estimator provides better goodness-of-fit between the observed and derived concentrations of water quality parameter, e.g., chlorophyll-a in turbid water. The kernel estimator is feasible for a relatively small size of ground observations. Approximately 30% improvement of cross-validation error was identified in this approach when compared with traditional regressions. The model offers a robust approach without further calibrations for estimating the spatial patterns of water quality by using remote sensing reflectance and a small set of observations, considering spatial and spectral information, e.g., multiple bands and band ratios.

Because of fast-paced industrialization, urbanization, and population growth in Indonesia, there are serious health issues in the country resulting from air pollution. This study uses geospatial modelling technologies, namely land-use regression (LUR), geographically weighted regression (GWR), and geographic and temporal weighted regression (GTWR) models, to assess variations in particulate matter (PM10) and nitrogen dioxide (NO2) concentrations in Surabaya City, Indonesia. This is the first study to implement spatiotemporal variability of air pollution concentrations in Surabaya City, Indonesia. To develop the prediction models, air pollution data collected from seven monitoring stations from 2010 to 2018 were used as dependent variables, while land-use/land cover allocations within a 250 m to 5000 m circular buffer range surrounding the monitoring stations were collected as independent variables. A supervised stepwise variable selection procedure was applied to identify the important predictor variables for developing the LUR, GWR, and GTWR models. The developed models of LUR, GWR, and GTWR accounted for 49%, 50%, and 51% of PM10 variations and 46%, 47%, and 48% of NO2 variations, respectively. The GTWR model performed better (R2 = 0.51 for PM10 and 0.48 for NO2) than the other two models (R2 = 0.49&ndash, 0.50 for PM10 and 0.46&ndash, 0.47 for NO2), LUR and GWR. In the PM10 model four predictor variables, public facility, industry and warehousing, paddy field, and normalized difference vegetation index (NDVI), were selected during the variable selection procedure. Meanwhile, paddy field, residential area, rainfall, and temperature played important roles in explaining NO2 variations. Because of biomass burning issues in South Asia, the paddy field, which has a positive correlation with PM10 and NO2, was selected as a predictor. By using long-term monitoring data to establish prediction models, this model may better depict PM10 and NO2 concentration variations within areas across Asia.

The retrieval of chlorophyll-a (Chl-a) concentrations relies on empirical or analytical analyses, which generally experience difficulties from the diversity of inland waters in statistical analyses and the complexity of radiative transfer equations in analytical analyses, respectively. Previous studies proposed the utilization of artificial neural networks (ANNs) to alleviate these problems. However, ANNs do not consider the problem of insufficient in situ samples during model training, and they do not fully utilize the spatial and spectral information of remote sensing images in neural networks. In this study, a two-stage training is introduced to address the problem regarding sample insufficiency. The neural network is pretrained using the samples derived from an existing Chl-a concentration model in the first stage, and the pretrained model is refined with in situ samples in the second stage. A novel convolutional neural network for Chl-a concentration retrieval called WaterNet is proposed which utilizes both spectral and spatial information of remote sensing images. In addition, an end-to-end structure that integrates feature extraction, band expansion, and Chl-a estimation into the neural network leads to an efficient and effective Chl-a concentration retrieval. In experiments, Sentinel-3 images with the same acquisition days of in situ measurements over Laguna Lake in the Philippines were used to train and evaluate WaterNet. The quantitative analyses show that the two-stage training is more likely than the one-stage training to reach the global optimum in the optimization, and WaterNet with two-stage training outperforms, in terms of estimation accuracy, related ANN-based and band-combination-based Chl-a concentration models.

Water is a significant factor in the occurrence of landslides when rainwater absorbs into the impermeable layer of soil and causes the soil to experience weathering; thus, that the soil is not strong enough to withstand the water load. Therefore, rainfall, as well as a land slope that affects the acceleration of runoff, becomes an important indicator that affects the potential for landslides. In this research, the effect of rainfall intensity on landslide was analyzed based on monthly data collected from 81 rain gauge stations. This idea was an effort to support research related to landslides mapping on the steep area. Thus, the analysis of potential landslides that vary according to rainfall changes monthly was obtained. La Nina phenomenon in 2016 causes the intensity of rainfall in the dry season to be quite high. However, the intensity values continued to show a significant change between dry and rainy seasons. The area of low landslide hazard changed significantly for low rainfall intensity (July 2016) and high intensity in April 2017. The total area of landslide hazard in April 2017 was 1.09 and 0.82 times more massive than in August 2016 for high and moderate potential hazards, respectively. Based on the characteristic of rainfall that change frequently and water as the primary trigger of the landslide, in determining the potential for landslides, rainfall factors need to be considered more detailed.

Processing of multitemporal satellite images generally suffers from uncertainties caused by differences in illumination and observation angles, as well as variation in atmospheric conditions. Moreover, satellite images acquired from different sensors contain not only the uncertainties but disparate relative spectral response. Given that radiometric calibration and correction of satellite images are difficult without ground measurements during data acquisition, this study addresses pseudo-invariant feature selection for relative radiometric normalization (RRN) that minimizes the radiometric differences among images caused by atmospheric and spectral band inconsistencies during data acquisition. The key to a successful RRN is the selection of pseudo-invariant features (PIFs) among bitemporal images. To select PIFs, multivariate alteration detection (MAD) algorithm is adopted with kernel canonical correlation analysis (KCCA) instead of canonical correlation analysis (CCA). KCCA, which assumes that the relation between at-sensor radiance is spatially nonlinear, can obtain more appropriate PIFs for cross-sensor images than that of CCA, which assumes that the relation between the at-sensor radiances of bitemporal image is spatially linear. In addition, a regularization term is added to the optimization of KCCA to avoid trivial solutions and overfitting. Qualitative and quantitative analyses on bitemporal images acquired by Landsat-7 Enhanced Thematic Mapper Plus and Landsat-8 Operational and Imager sensors were conducted to evaluate the proposed method. The experimental results demonstrate the superiority of the proposed KCCA based MAD to the CCA-based MAD in terms of PIF selection, particularly for images containing significant cloud