Michel Castro Moreira

Continuous streamflow monitoring is fundamental for decision-making in water resources management. However, direct streamflow measurement is costly, complex, and laborious. This paper presents ARCADE, a software for hydraulic diagnostics under different flow conditions, rating curve fitting, and discharge determination. It integrates hydrological modeling, statistical metrics, and outlier detection using R. ARCADE features a flexible architecture that adapts to different geographical contexts, datasets, and operational assumptions, supporting hydrological research and experimentation.

Rainfall time series are essential for hydrological and climate studies; however, data scarcity remains a persistent challenge that compromises the reliability of analyses and modeling. This study compares the performance of regression based models and machine learning (ML) methods for gap filling in monthly rainfall data from Northern Minas Gerais, Brazil, a region characterized by high climatic variability and limited monitoring infrastructure. Ten missing data levels, ranging from 5% to 50%, were simulated using historical records from seven rainfall stations. Four regression based approaches, Regional Weighting (RW), Simple Linear Regression (SLR), Multiple Linear Regression (MLR), and Regression Based Weighting (RWR), were evaluated alongside three ML methods, Random Forest (RF), Support Vector Machines (SVM), and k Nearest Neighbors (KNN). Model performance was assessed using Root Mean Square Error (RMSE), Symmetric Mean Absolute Percentage Error (SMAPE), Nash Sutcliffe Efficiency (NSE), and the coefficient of determination (R ² ), with nonparametric statistical tests applied to identify significant differences. The results showed that regression and weighting based models consistently outperformed ML methods across all missing data levels. RWR and RW achieved the best overall performance, with the lowest RMSE and SMAPE values and the highest NSE and R ² values, and no statistically significant differences were observed between them. MLR performed well only at moderate levels of missing data, with significant performance degradation beyond 40% of missing values. Among ML methods, RF and SVM showed intermediate performance, while KNN and SLR yielded the weakest results. These findings highlight that weighting schemes based on the relevance and proximity of neighboring stations provide more robust rainfall estimates than complex nonlinear models in data scarce and highly variable climatic regions. Graphical Abstract The graphical abstract presents the workflow used to evaluate the performance of gap-filling techniques in monthly rainfall time series from Northern Minas Gerais, Brazil. The process begins with historical rainfall data, in which artificial gaps ranging from 5% to 50% are systematically introduced to simulate missing values. These incomplete time series are then processed using seven gap filling models that include both statistical and machine learning approaches, namely Regional Weighting (RW), Simple Linear Regression (SLR), Multiple Linear Regression (MLR), Regression Based Weighting (RWR), Random Forest (RF), Support Vector Machines (SVM), and K Nearest Neighbors (KNN). Each technique is applied iteratively across 100 simulations to ensure statistical reliability. The performance of the reconstructed series is evaluated using four metrics, Root Mean Square Error (RMSE), Symmetric Mean Absolute Percentage Error (SMAPE), Nash Sutcliffe Efficiency (NSE), and the coefficient of determination (R ² ). The flowchart visually synthesizes this stepwise approach, covering data preprocessing, gap simulation, model application, and performance comparison, and highlights the systematic strategy adopted in the study to identify the most accurate gap filling models under different levels of missing data.

Coffee production in Brazil requires efficient monitoring tools, especially in regions with heterogeneous terrain and planting patterns. Accurate delineation of coffee canopies from UAV imagery is essential for extracting reliable agronomic information and supporting precision agriculture practices. Multispectral UAV images were acquired over a commercial coffee farm in the Matas de Minas region, Brazil. An orthomosaic was generated and manually annotated to create binary ground-truth masks of coffee canopies. A U-Net convolutional neural network was trained using multispectral bands and the Normalized Difference Vegetation Index (NDVI) as input. Model performance was evaluated using Dice coefficient, Intersection over Union (IoU), precision, recall, and accuracy. The proposed model achieved a Dice coefficient of 0.89, IoU of 0.83, and overall accuracy above 0.92. The predicted coffee canopy area (1.50 ha) was consistent with the reference area obtained from manual annotation (1.45 ha), indicating reliable canopy delineation under heterogeneous field conditions. The results demonstrate that semantic segmentation using multispectral UAV imagery and a U-Net model is an effective and applicable approach for coffee canopy mapping, with potential use in precision agriculture and farm-scale decision-support systems.

Precipitation is the primary source of water in a river basin and plays a key role in water resource management by affecting water availability. However, large river basins like the São Francisco River Basin in Brazil show significant variability in rainfall patterns, leading to different water supply levels across sub-regions. To identify areas with similar rainfall patterns, the k-means clustering method was used on rainfall stations throughout the basin. The clusters were validated by analyzing their means, medians, and interquartile ranges. This analysis was supported by a multi-criteria assessment using two methods: one with a correlation matrix and the other with a covariance matrix. As a result, the stations were grouped into five homogeneous clusters based on precipitation. The two clusters in the southern basin were heavily influenced by the South Atlantic Convergence Zone (SACZ). In contrast, the northern clusters were mainly affected by the Intertropical Convergence Zone (ITCZ) and orographic features such as the Borborema Plateau. The coastal cluster was influenced by the sea and trade winds blowing from east to west. Multivariate analysis showed high correlation and seasonal similarity among stations within the same cluster. Clusters 4 and 5 experienced the highest average precipitation, while clusters 1 and 2 recorded the lowest, with cluster 2 being particularly notable.

This paper aimed to perform an ecohydrological analysis of the Ondas river basin, located in the Brazilian Cerrado, using the Physical Habitat Simulation (PHABSIM) model, in order to determine the monthly ecological flows for its lower course. For calibration and simulation in PHABSIM, field experiments were conducted during dry and rainy periods, in a 300‐m stretch located in the lower course of the Ondas river. The estimate of monthly ecological flows was obtained by analysing the Weighted Usable Area of the ichthyofauna species of the river stretch, as a function of the streamflows with a probability of nonexceedance of 50% (Q₅₀), 60% (Q₆₀), 70% (Q₇₀), 80% (Q₈₀), 90% (Q₉₀) and 95% (Q₉₅). In the dry period, the ecological flow varied between 31.72 and 40.14 m³ s⁻¹, whereas in the rainy season, it presented values between 33.23 and 51.94 m³ s⁻¹. Considering the water use rights criterion of the State of Bahia, Brazil, where it is permissible to use up to 80% of the Q₉₀, it was verified that the adoption of ecological flows would considerably reduce the quantity of water that could be used for human purposes but would maintain the habitat of the bioindicator species studied. The ecological flow regime obtained provides subsidies for discussions and negotiations on the water resources management in the Ondas river basin, considering the ecohydrological aspects affecting the region, in addition to the water quantitative and qualitative factors.

Rainfall erosivity (RE) represents the potential of rainfall to cause soil erosion, and understanding its impact is essential for the adoption of soil and water conservation practices. Although several studies have estimated RE for Brazil, currently, no single reliable and easily accessible database exists for the country. To fill this gap, this work aimed to review the research and generate a rainfall erosivity database for Brazil. Data were gathered from studies that determined rainfall erosivity from observed rainfall records and synthetic rainfall series. Monthly and annual rainfall erosivity values were organized on a spreadsheet and in the shapefile format. In total, 54 studies from 1990 to 2023 were analyzed, resulting in the compilation of 5516 erosivity values for Brazil, of which 6.3% were pluviographic, and 93.7% were synthetic. The regions with the highest availability of information were the Northeast (35.6%), Southeast (30.1%), South (19.9%), Central-West (7.7%), and North (6.7%). The database, which can be accessed on the Mendeley Data platform, can aid professionals and researchers in adopting public policies and carrying out studies aimed at environmental conservation and management basin development.

Vegetation indices have recently been proposed for remote sensing SAR (synthetic aperture radar) sensor measurements to monitor vegetation. However, they still lack validation studies on different vegetation types for their correct application. Thus, the objective of this study was to test the applicability of the Dual-polarization SAR Vegetation Index (DPSVI) and the modified DPSVI (DPSVIm) as indicators of aboveground biomass (AGB) from dense forest fragments.

Despite optical remote sensing (and the spectral vegetation indices) contributions to digital soil-mapping studies of soil organic carbon (SOC), few studies have used active radar remote sensing mission data like that from synthetic aperture radar (SAR) sensors to predict SOC. Bearing in mind the importance of SOC mapping for agricultural, ecological, and climate interests and also the recently developed methods for vegetation monitoring using Sentinel-1 SAR data, in this work, we aimed to take advantage of the high operationality of Sentinel-1 imaging to test the accuracy of SOC prediction at different soil depths using machine learning systems. Using linear, nonlinear, and tree regression-based methods, it was possible to predict the SOC content of soils from western Bahia, Brazil, a region with predominantly sandy soils, using as explanatory variables the SAR vegetation indices. The models fed with SAR sensor polarizations and vegetation indices produced more accurate results for the topsoil layers (0–5 cm and 5–10 cm in depth). In these superficial layers, the models achieved an RMSE in the order of 5.0 g kg⁻¹ and an R² ranging from 0.16 to 0.24, therefore explaining about 20% of SOC variability using only Sentinel-1 predictors.