Globally, Smallholder farming systems (SFS) are recognized as one of the most important pillars of rural economic development and poverty alleviation because of their contribution to food security. However, support for this agricultural sector is hampered by lack of reliable information on the distributions and acreage of smallholder fields. This information is essential in not only monitoring food security and informing markets but also in guiding the determination of levels of support required from government by individual farmers. There is urgent need for robust techniques that can be used to cost-effectively and time-efficiently map smallholder crop fields especially in Sub-Saharan Africa and Asia. This study attempts to do this by using an approach in which optical and Synthetic Aperture Radar (SAR) data are systematically combined and classified using Extreme Gradient Boosting (Xgboost). We also investigated model stacking as another technique to improve classification accuracy. We combined Xgboost with Random Forest (RF), Support Vector Machine (SVM), Artificial Neural Networks (ANN), and Naïve Bayes (NB). The combined use of multi-temporal Sentinel-2 bands, spectral indices, and Sentinel-1 produced better results than exclusive use of optical data (α = 0.95, p = 0.0005). Furthermore, stacking of classification algorithms based on model comparisons achieved higher accuracy than stacking the algorithms indiscriminately (α = 0.95, p = 0.0100). Through systematic fusion of SAR and optical data and hyper-parameter tuning of Xgboost, we achieved a maximum classification accuracy of 97.71%, while achieving a maximum accuracy of 96.06% through model stacking. This highlights the importance of multi-sensor data fusion and multi-classifier systems when mapping fragmented agricultural landscapes.

Precise adjustments of farm management activities, such as irrigation and soil treatment according to site-specific conditions, are crucial. With advances in smart agriculture and sensors, it is possible to reduce the cost of water and soil treatment inputs but still realize optimal yields and high-profit returns. However, achieving precise application requirements cannot be efficiently practiced with spatially disjointed information. This study assessed the potential of using an electromagnetic induction device (EM38-MK) to cover this gap. An EM38-MK was used to measure soil apparent electrical conductivity (ECa) as a covariate to determine soil salinity status and soil water content θ post irrigation at four depth layers (Hz: 0–0.25 m; Hz: 0–0.75 m; Vz: 0.50–1 m). The inverse distance weighting method was used to generate the spatial distribution thematic layers of electrical conductivity. The statistical measures showed an R² = 0.87; r > 0.7 and p ≤ 0.05 on correlation of ECa and SWC. Based on the South African salinity class of soils, the area was not saline ECa < 200 mS/m. The EM38-MK can be used to estimate soil salinity and SWC variability using ECa as a proxy, allowing precise estimations with depths and in space. These findings provide key information that can aid in irrigation scheduling and soil management.

We evaluated unified algorithms for remote sensing of chlorophyll-a (Chla) and turbidity in eutrophic and ultra-turbid waters such as Japan’s Lake Shinji and Lake Nakaumi (SJNU) and the Vaal Dam Reservoir (VDR) in South Africa. To realize this objective, we used 38 remote sensing reflectance (R_rs), Chla and turbidity datasets collected in these waters between July 2016 and March 2017. As a result, we clarified the following items. As a unified Chla model, we obtained strong correlation (R² = 0.7, RMSE = 2 mg m⁻³) using a two-band model (2-BM) and three-band model (3-BM), with R_rs(687)/R_rs(672) and [R_rs⁻¹(687) − R_rs⁻¹(672)] × R_rs(832). As a unified turbidity model, we obtained strong correlation (R² = 0.7, RMSE = 260 NTU) using 2-BM and 3-BM, with R_rs(763)/R_rs(821) and R_rs(810) − [R_rs(730) + R_rs(770)]/2. When targeting the Sentinel-2 Multispectral Imager (MSI) frequency band, we focused on MSI Bands 4 and 5 (R_rs(740) and R_rs(775)) for the Chla algorithm. When optically separating SJNU and VDR data, it is effective to use the slopes of MSI Bands 3 and 4 (R_rs(560) and R_rs(665)) and the slopes of MSI Bands 7 and 9 (R_rs(775) and R_rs(865)).

Reducing food insecurity in developing countries is one of the crucial targets of the Sustainable Development Goals (SDGs). Smallholder farmers play a crucial role in combating food insecurity. However, local planning agencies and governments do not have adequate spatial information on smallholder farmers, and this affects the monitoring of the SDGs. This study utilized Sentinel-1 multi-temporal data to develop a framework for mapping smallholder maize farms and to estimate maize production area as a parameter for supporting the SDGs. We used Principal Component Analysis (PCA) to pixel fuse the multi-temporal data to only three components for each polarization (vertical transmit and vertical receive (VV), vertical transmit and horizontal receive (VH), and VV/VH), which explained more than 70% of the information. The Support Vector Machine (SVM) and Extreme Gradient Boosting (Xgboost) algorithms were used at model-level feature fusion to classify the data. The results show that the adopted strategy of two-stage image fusion was sufficient to map the distribution and estimate production areas for smallholder farms. An overall accuracy of more than 90% for both SVM and Xgboost algorithms was achieved. There was a 3% difference in production area estimation observed between the two algorithms. This framework can be used to generate spatial agricultural information in areas where agricultural survey data are limited and for areas that are affected by cloud coverage. We recommend the use of Sentinel-1 multi-temporal data in conjunction with machine learning algorithms to map smallholder maize farms to support the SDGs.

The goal of this study was to understand land use and land cover (LULC) changes within the lower uMfolozi floodplain system, South Africa, and relate those changes to wetland loss. Changes in LULC were assessed using a geographic object-based image analysis (GEOBIA) algorithm to classify multi-date Landsat images into eight cover types over a period of 20 years, between 1997 and 2017. Post-classification accuracy assessment of all map-outputs was conducted by compiling confusion matrixes and calculating producer, user, and global accuracies and kappa coefficients (K) for each map-output. Levels of accuracy for all map-outputs were within acceptable limits, ranging between 79% and 88% (K = 0.76 and 0.86, respectively). Thereafter, paired t-tests were applied to determine whether the changes in LULC over the study period were significant. Results of this investigation showed a significant (p-value, < 0.01) conversion of wetland to cultivation, by 14%. This finding is important because it demonstrates that in this environment, human agency is one of the major drivers of a persistent decrease in the wetland ecosystem. The major insight from this observation is that there is an urgent need to formulate and implement objectively informed interventions to enhance the sustainability of the uMfolozi floodplain system and that of others elsewhere.