Ulaanbaatar, the capital of Mongolia, has expanded rapidly over the past decade. Insufficient authority is in place to address this expansion, and many residential plots have been developed in the peripheral regions of the city. The aim of this study is to estimate changes in land cover within the central part of Ulaanbaatar, which has been affected by anthropogenic disturbances. The breaks for additive seasonal and trend (BFAST) method is a powerful tool for implementing this study because it is able to robustly and automatically derive the timing and locations of land cover changes from spatio-temporal datasets. We applied the BFAST method for the first time to urban expansion analysis, with NDVI time series calculated from MODIS (MOD09A1 product) during the period 2000–2010. The results show that land cover has changed across 22.51% of the study area, and that the change occurs at a later time with increasing distance from the city center. Bi-temporal high-resolution satellite images of a sample area in 2000 and 2008 confirmed that the detection of land cover changes by BFAST corresponds to areas in which residential development is dominant. This study demonstrates that BFAST is an effective method for monitoring urban expansion. In addition, it increases the applicability of NDVI time series.

Sago palm (Metroxylon sagu) is a palm tree species originating in Indonesia. In the future, this starch-producing tree will play an important role in food security and biodiversity. Local governments have begun to emphasize the sustainable development of sago palm plantations; therefore, they require near-real-time geospatial information on palm stands. We developed a semi-automated classification scheme for mapping sago palm using machine learning within an object-based image analysis framework with Pleiades-1A imagery. In addition to spectral information, arithmetic, geometric, and textural features were employed to enhance the classification accuracy. Recursive feature elimination was applied to samples to rank the importance of 26 input features. A support vector machine (SVM) was used to perform classifications and resulted in the highest overall accuracy of 85.00% after inclusion of the eight most important features, including three spectral features, three arithmetic features, and two textural features. The SVM classifier showed normal fitting up to the eighth most important feature. According to the McNemar test results, using the top seven to 14 features provided a better classification accuracy. The significance of this research is the revelation of the most important features in recognizing sago palm among other similar tree species.

The spatial plan program for Makassar City and the surrounding area called Mamminasata (Makassar, Maros, Sungguminasa, and Takalar) was created by the Indonesian Government. The program regulates the proportion of land cover, but predictions about land cover changes were not considered. Therefore, in this study, we predict what the land cover may be in 2031 using the multi-layer perceptron neural network and the Markov chain methods. For this purpose, image composite, support vector machine classifier, and change detection were applied to a time series of satellite data. Visual validation showed the hot-spots of land cover changes related to population density, and statistical validation scored 0.99 and 0.78 in no information kappa and grid-cell level location kappa, respectively. The model was performed to predict land cover in 2031, and the predicted result was then compared with the spatial plan using an overlapping method. The results showed that built-up area, dryland agriculture, and wetland agriculture occupied two, twenty, and eight percent of the protected zone, respectively. Meanwhile, fifteen percent of the development zone was covered by forest, mainly in the eastern part of Mamminasata. The result can be used to help the Government decide future plans for the Mamminasata area.

Vegetation coverage is a crucial parameter in agriculture, as it offers essential insight into crop growth and health conditions. The spatial resolution of spaceborne sensors is limited, hindering the precise measurement of vegetation coverage. Consequently, fine-resolution ground observation data are indispensable for establishing correlations between remotely sensed reflectance and plant coverage. We estimated rice plant coverage per pixel using time-series Sentinel-2 Multispectral Instrument (MSI) data, enabling the monitoring of rice growth conditions over a wide area. Coverage was calculated using unmanned aerial vehicle (UAV) data with a spatial resolution of 3 cm with the spectral unmixing method. Coverage maps were generated every 2–3 weeks throughout the rice-growing season. Subsequently, crop growth was estimated at 10 m resolution through multiple linear regression utilizing Sentinel-2 MSI reflectance data and coverage maps. In this process, a geometric registration of MSI and UAV data was conducted to improve their spatial agreement. The coefficients of determination (R²) of the multiple linear regression models were 0.92 and 0.94 for the Level-1C and Level-2A products of Sentinel-2 MSI, respectively. The root mean square errors of estimated rice plant coverage were 10.77% and 9.34%, respectively. This study highlights the promise of satellite time-series models for accurate estimation of rice plant coverage.

Himawari-8/AHI is a new geostationary sensor that can observe the land surface with high temporal frequency. Bidirectional reflectance derived by the Advanced Himawari Imager (AHI) includes information regarding land surface properties such as albedo, vegetation condition, and forest structure. This information can be extracted by modeling bidirectional reflectance using a bidirectional reflectance distribution function (BRDF). In this study, a kernel-driven BRDF model was applied to the red and near infrared reflectance observed over 8 hours during daytime to express intraday changes in reflectance. We compared the goodness of fit for six combinations of model kernels. The Ross-Thin and Ross-Thick kernels were selected as the best volume kernels for the red and near infrared bands, respectively. For the geometric kernel, the Li-sparse-Reciprocal and Li-Dense kernels displayed similar goodness of fit. The coefficient of determination and regression residuals showed a strong dependency on the azimuth angle of land surface slopes and the time of day that observations were made. Atmospheric correction and model adjustment of the terrain were the main issues encountered. These results will help to improve the BRDF model and to extract surface properties from bidirectional reflectance.

Correction of spectral reflectance for shadow and topography in optical remote sensing data is challenging. Here, we corrected for canopy self-shadowing in an evergreen conifer forest in mountainous terrain using a three-dimensional (3D) point cloud. In our approach, the surface was modeled from structure-from-motion processed images provided by an unmanned aerial vehicle; then, the relationship between the observed spectral reflectance of the Sentinel-2A/B multispectral instrument, and the simulated sunlit fraction (the percentage of the solar-illuminated area within a Sentinel-2 pixel) was determined based on a ray-tracing scheme using a 3D point cloud. Scene-based and pixel-based linear regressions were applied to remove canopy-shadow and topographic effects from satellite-observed reflectance. Scene-based regression resulted in large seasonal changes that caused overcorrection in winter. Pixel-based regression generated stable seasonal profiles in both the red and near-infrared reflectance values over the conifer canopy; however, excessively smoothed seasonal changes were implemented over deciduous vegetation. In both correction methods, the reflection of incident light by the canopy likely improved the correction accuracy by decreasing the contrast between illuminated and shaded pixels in summer and increasing it in winter. The results were also visually compared with those from the Sun-Canopy-Sensor with C (SCS+C) method and the Sentinel-2 Level-2A product. This research demonstrates the effectiveness of using a 3D point cloud to correct for self-shadowing and topographic effects on remotely sensed reflectance.