In previous classification studies, three non-parametric classifiers, Random Forest (RF), k-Nearest Neighbor (kNN), and Support Vector Machine (SVM), were reported as the foremost classifiers at producing high accuracies. However, only a few studies have compared the performances of these classifiers with different training sample sizes for the same remote sensing images, particularly the Sentinel-2 Multispectral Imager (MSI). In this study, we examined and compared the performances of the RF, kNN, and SVM classifiers for land use/cover classification using Sentinel-2 image data. An area of 30 × 30 km² within the Red River Delta of Vietnam with six land use/cover types was classified using 14 different training sample sizes, including balanced and imbalanced, from 50 to over 1250 pixels/class. All classification results showed a high overall accuracy (OA) ranging from 90% to 95%. Among the three classifiers and 14 sub-datasets, SVM produced the highest OA with the least sensitivity to the training sample sizes, followed consecutively by RF and kNN. In relation to the sample size, all three classifiers showed a similar and high OA (over 93.85%) when the training sample size was large enough, i.e., greater than 750 pixels/class or representing an area of approximately 0.25% of the total study area. The high accuracy was achieved with both imbalanced and balanced datasets.

Losses of biodiversity and ecosystem functioning due to rainforest destruction and agricultural intensification are prime concerns for science and society alike. Potentially, ecosystems show nonlinear responses to land-use intensification that would open management options with limited ecological losses but satisfying economic gains. However, multidisciplinary studies to quantify ecological losses and socioeconomic tradeoffs under different management options are rare. Here, we evaluate opposing land use strategies in cacao agroforestry in Sulawesi, Indonesia, by using data on species richness of nine plant and animal taxa, six related ecosystem functions, and on socioeconomic drivers of agroforestry expansion. Expansion of cacao cultivation by 230% in the last two decades was triggered not only by economic market mechanisms, but also by rarely considered cultural factors. Transformation from near-primary forest to agroforestry had little effect on overall species richness, but reduced plant biomass and carbon storage by ≈75% and species richness of forest-using species by ≈60%. In contrast, increased land use intensity in cacao agroforestry, coupled with a reduction in shade tree cover from 80% to 40%, caused only minor quantitative changes in biodiversity and maintained high levels of ecosystem functioning while doubling farmers' net income. However, unshaded systems further increased income by ≈40%, implying that current economic incentives and cultural preferences for new intensification practices put shaded systems at risk. We conclude that low-shade agroforestry provides the best available compromise between economic forces and ecological needs. Certification schemes for shade-grown crops may provide a market-based mechanism to slow down current intensification trends.

Recently, several methods have been introduced and applied to estimate daily air surface temperature (T_a) using MODIS land surface temperature data (MODIS LST). Among these methods, the most common used method is statistical modeling, and the most applied algorithms are linear/multiple linear regression models (LM). There are only a handful of studies using machine learning algorithm models such as random forest (RF) or cubist regression (CB). In particular, there is no study comparing different combinations of four MODIS LST datasets with or without auxiliary data using different algorithms such as multiple linear regression, random forest, and cubist regression for daily T_a-max, T_a-min, and T_a-mean estimation. Our study examines the mentioned combinations of four MODIS-LST datasets and shows that different combinations and differently applied algorithms produce various T_a estimation accuracies. Additional analysis of daily data from three climate stations in the mountain area of North West of Vietnam for the period of five years (2009 to 2013) with four MODIS LST datasets (AQUA daytime, AQUA nighttime, TERRA daytime, and TERRA nighttime) and two additional auxiliary datasets (elevation and Julian day) shows that CB and LM should be applied if MODIS LST data is used solely. If MODIS LST is used together with auxiliary data, especially in mountainous areas, CB or RF is highly recommended. This study proved that the very high accuracy of T_a estimation (R² > 0.93/0.80/0.89 and RMSE ~1.5/2.0/1.6 °C of T_a-max, T_a-min, and T_a-mean, respectively) could be achieved with a simple combination of four LST data, elevation, and Julian day data using a suitable algorithm.

Statistical independence test and validity of the CA (Cellular Automata) Markov process for projecting future land use and land cover (LULC) changes were carried out in this study. Predicting quantity and location changes have been analyzed, and statistically evaluated. Validity of the CA Markov process has been examined using various Kappa Index of Agreement (KIA or Kstandard) and related statistical variations on the KIA. Statistical test of independence (K2) was performed and markovian suitability has been checked using hypothesis of goodness of fit (Xc2). Hypothesis of statistical independence was rejected, which proved that land use land cover change trends are similar like previous development of land. With acceptance of the hypothesis of goodness of fit (Xc2) proved that actual transition probability of matrix is fitted with expected transition probability prepared using Markov chain method. Statistics indicates Kno, Klocation, Klocation Strata and Kstandard are 0.8347, 0.859, 0.8591 and 0.7928, respectively

Soil property maps are essential resources for agricultural land use. However, soil properties mapping is costly and time-consuming, especially in the regions with complicated topographic conditions. This study was conducted in a hilly region of Central Vietnam with the following objectives: (i) to evaluate the best environmental variables to estimate soil organic carbon (SOC), total nitrogen (TN), and soil reaction (pH) with a regression kriging (RK) model, and (ii) to compare the accuracy of the ordinary kriging (OK) and RK methods. SOC, TN, and soil pH data were measured at 155 locations within the research area with a sampling grid of 2 km × 2 km for a soil layer from 0 to 30 cm depth. From these samples, 117 were used for interpolation, and the 38 randomly remaining samples were used for evaluating accuracy. The chosen environmental variables are land use type (LUT), topographic wetness index (TWI), and transformed soil adjusted vegetation index (TSAVI). The results indicate that the LUT variable is more effective than TWI and TSAVI for determining TN and pH when using the RK method, with a variance of 7.00% and 18.40%, respectively. In contrast, a combination of the LUT and TWI variables is the best for SOC mapping with the RK method, with a variance of 14.98%. The OK method seemed more accurate than the RK method for SOC mapping by 3.33% and for TN mapping by 10% but the RK method was found more precise than the OK method for soil pH mapping by 1.81%. Further selection of auxiliary variables and higher sampling density should be considered to improve the accuracy of the RK method.

Land evaluation is a process that is aimed at the sustainable development of agricultural production in rural areas, especially in developing countries. Therefore, land evaluation involves many aspects of natural conditions, economic, and social issues. This research was conducted in a hilly region of Central Vietnam to assess the land suitability of potential agricultural land use types that are based on scientific and local knowledge. In the frame of this research, Participatory Rural Appraisal (PRA); Analytical Hierarchy Analysis (AHP); Geographic Information System (GIS); and, scoring based scientific literature and local knowledge were applied for Multi-Criteria Decision Analysis (MCDA) for land use evaluation. The results of the PRA survey reveal that five plants offer great agricultural potential in the research area, namely rice, cassava, acacia, banana, and rubber. The land suitability of each plant type varies, depending on physical conditions as well as economic and social aspects. Acacia and cassava represent the most suitable plant types in the research area. Recommendations regarding agricultural land use planning in the A Luoi district are brought forward based on the land evaluation results. The combination of scientific and local knowledge in land assessment based on GIS technology, AHP, and PRA methods is a promising approach for land evaluation.