Gauge Stations Research Articles

Hydrological Modeling of Krishna Upper Catchment Area of India Using Multisite Calibration and Validation of SWAT Model

Multisite calibration & validation of model reduces the uncertainty present in the model parameters and improve the SWAT model output. In the present study, an effort was made to set up SWAT model along with multisite calibration and validation for Kishna upper basin of India. A total of six gauging stations maintained by Central water commission across the area were selected for SWAT model calibration and validation. Sequential Uncertainty Fitting (SUFI-2) programme of SWAT-CUP (Calibration and Uncertainty Programme) was employed to calibrate the SWAT model. During sensitivity analysis of SWAT model, parameters which are highly sensitive that influences flow characteristics of different gauging station were identified. These parameters along with parameter range were used to calibration and validation of SWAT model each of the sub basins. Observed flow and predicted stream flows were compared with certain model performance indices. These indices indicated that for all six gauging stations, the SWAT model found to be very good. The performance indices like R2, NSE, RSR and PBIAS were ranged from 0.70 to 0.84, 0.67 to 0.76, 0.49 to 0.57 and -11.7 to 24.0% respectively during model calibration period and 0.70 to 0.93, 0.67 to 0.83, 0.41 to 0.58 and 8.5 to 22.9% respectively during model validation period. A good agreement between observed and simulated flows at all the gauging stations were observed. It can be concluded that SWAT model output was improved with multisite calibration and validation of the model.

Chronological changes in the Black Sea level at tide gauge station Varna 1928-2023

This paper outlines time series data analysis from Varna tidal station (Black sea, Bulgaria), with the objective being to determine the influences of long-period and short-period tides. Observations series from monthly mean sea level values cover the period 1928-2023. The sea level measurements until 2013 were done with a mechanical float tide gauge. Since 2013 a radar tide gauge was installed in place of the mechanical one. A least squares linear regression model was applied for the analysis of monthly sea level data. As a result of the analysis, mean sea level, trend, seasonal and long-term tidal influences were established. In order to determine the influence of short-term tides, the method of least squares harmonic analysis was applied. In this method, observed sea level values are considered to be represented as a sum of tidal harmonics operating at previously known frequencies. The observation data include hourly values for the period 06.2013-12.2023. Some of the advantages of the applied method are that it allows to analyze series with interruptions in the observations, as well as assist to easily determine tides with close frequencies. From the analysis was determined the influence of: the annual and semi-annual tides, as well as about 68 short-period tides.

Comprehensive comparative analysis of reconstructed sea level datasets in the China Seas: insights from tide gauge and satellite altimetry

At present, there are many reconstructed datasets at the global scale. To test the applicability of these datasets in the China seas, the study comprehensively analyzes the reliability and accuracy of reconstructed sea level datasets in capturing nuanced temporal patterns of sea level changes in the China Seas. This study applied analysis methods or indicators such as time series, Taylor plots, correlation coefficients, growth rates, and standard deviations. Ocean Data Assimilations (ODAs) outperform Tide Gauge Reconstructions (TGRs) in terms of correlation with measured data in the nearshore, while TGRs exhibit superior capability in capturing oceanic sea level variability. Although the ODAs and TGRs both suffer from the underestimation of sea level variability in China as well as in neighboring seas, the TGRs perform better than the former. ODAs show inconsistency in reflecting the rate of sea level rise, but they, particularly the China Ocean Reanalysis (CORA), demonstrate a better correlation with satellite altimetry datasets. Meanwhile, both of them can reflect the Pacific Decadal Oscillation (PDO) well. TGRs, relying on oceanic tide gauge stations, suffer from poor correlation with tide gauge stations due to limited coverage. Reconstruction discrepancies are attributed to methodological differences and data assimilation techniques. Future studies should explore alternative variables like sea surface temperature and so on to enhance sea-level reconstruction, especially in regions with sparse tide gauge coverage.

Downscaling MODIS evapotranspiration into finer resolution using machine learning approach on a small scale, Ribb watershed, Ethiopia.

By monitoring evapotranspiration (ET), the exchange of water and energy between the soil, plants, and the atmosphere can be controlled. Routine estimations of ET on a daily, monthly, and seasonal basis can give relevant information on small-scale agricultural practices, such as the Ribb watershed in Ethiopia. However, MODIS sensors have recently given high temporal resolution ET products across large areas, but their low spatial resolution limits its application on a local scale. The primary goal of the study was to downscale the MODIS ET (1km) product to a finer spatial resolution at the watershed level. The model's 12 predictor variables (NDVI, EVI, LAI, FVC, SAVI, NDMI, NDWI, Albedo, emissivity, LST, and DEM: slope and elevation) were produced using the random forest (RF) algorithm using Sentinel-2 (S-2) 20m and Landsat-8 (L-8) 30m. The RF regression model was used to assess the relationship between predicted variables and downscaled MODIS ET. The FAO-PM ET model, developed from meteorological stations, was validated by and RMSE for three seasons (rainy, post-rainy, and dry) in 2022. The results were in good agreement with MODIS ET, with an RMSE of 0.22 for S-2 and 0.28 for L-8. In the FAO-PM ET model, the downscaled result showed greater spatial details and better agreement with gage station readings ( and ). Thus, considering the effectiveness and simplicity of machine learning techniques, our study demonstrated the potential for ET downscaling. Furthermore, the study suggests integrating spatiotemporal time series data to reach higher resolution.

A Stepwise‐Clustered Precipitation Downscaling Method for Ensemble Climatic Projections in the Mediterranean Region

ABSTRACTPrecipitation changes dynamically in the Mediterranean region. Therefore, the projection of future precipitation and its historical distribution mechanism is essential for climate mitigation and adaptation. In this study, a stepwise clustered precipitation downscaling method (SCPD) was developed and adopted in the Mediterranean region to reveal the inherent variation rules and trends over the future 100 years under two SSP scenarios. A cutting and merging multivariate process is introduced to build a cluster tree for supporting further downscaling and projecting steps. The ensemble average from the global climate model (GCM) dataset is used for precipitation projections. The precipitation performance of SCPD, evaluated by R2, is fairly decent. The precipitation projections vary with the original rainfall patterns over the gauge stations. Dry places tend to become comparably drier in the future. Precipitation in the northern Mediterranean region shows a drier winter–spring and wetter summer–autumn. Opposite trends emerged in the southern part, with increasing winter precipitation and decreasing summer rainfall. The rising carbon dioxide concentration will further intensify the decrease in rainfall. However, the centres of these two EOFs are not identical. The contributions of NAO (positive) and Niño 3.4 (negative) to PC1 are relatively high. Accordingly, the strongest positive correlation with PC2 is SCAND, as well as negative correlations with AO, NAO and EAWR. Positive anomaly precipitation is attributed to PC1, whereas PC2 is responsible for most of the negative variance precipitation.

Multi-Step Ahead Water Level Forecasting Using Deep Neural Networks

Stream gauge height (water level) is a significant indicator for forecasting future floods. Flooding occurs when the water level exceeds the flood stage. Predicting imminent floods can save lives, protect infrastructure, and improve road traffic management and transportation. Deep neural networks have been increasingly used in this domain due to their predictive capabilities in capturing complex features and interdependencies. This study employs four distinct models—Multi-Layer Perceptron (MLP), Long Short-Term Memory (LSTM), transformer, and LSTNet—with MLP serving as the baseline model to forecast water levels. The models are trained using data from 20 distinct river gages across the state of Missouri to ensure consistent performance. Random search optimization is employed for hyperparameter tuning. The prediction intervals are set at 4, 6, 8, and 10 (each interval equivalent to 30 min) to ensure that performance results are robust and not due to random weight initialization or suboptimal hyperparameters and are consistent throughout different prediction intervals. The findings of this study indicate that the LSTNet model leads to a better performance than the other models, with a median RMSE of 0.00724, 0.00959, 0.01204, and 0.01230 for the 4, 6, 8, and 10 intervals, respectively. As climate change leads to localized storms driven by atmospheric shifts, water level fluctuations are becoming increasingly extreme, further exacerbating data drift in real-world datasets. The LSTNet model demonstrates superior performance in terms of RMSE, MAE, and the correlation coefficient across all prediction intervals when forecasting water levels under data drift conditions.

Análise de cluster e regionalização hidrológica para Estados brasileiros

ABSTRACT Streamflow data from gauging stations are essential for effective water resources management. However, some regions in Brazil lack the necessary data. Hydrological regionalization is an alternative technique for obtaining data such regions. However, not all regions in Brazil have defined hydrological regionalization models, including the state of Goiás and the Brazilian Federal District. The objective of this study was to develop a hydrological regionalization methodology based on the separation of hydrologically homogeneous regions and multiple linear regression, using a Geographic Information System (GIS) program. Historical series data were used to calculate reference flows with 90 or 95% duration over time in the watercourse (Q90 and Q95) and the mean flow (Ǭ). Rain gauge station data were used to calculate the mean annual rainfall in each watershed through spatial interpolation by ordinary kriging. Subsequently, the physiographic characteristics of each watershed were calculated. The hydrologically homogeneous regions were delimited based on these data using cluster analysis, which identified seven hydrologically homogeneous regions in Goiás, two of them belonging to the Federal District. Multiple regression allowed the development of seven regionalization models. Models for regions 1, 3, 4, 5, and 7 showed better performance.

Enhancing the performance of runoff prediction in data-scarce hydrological domains using advanced transfer learning

Accurate hydrological predictions are often hindered by the lack of stream gauges in data-scarce regions, where traditional transfer learning (TL) models like Long Short-Term Memory (LSTM) networks often face limitations due to reduced accuracy and adaptability. To enhance runoff prediction in such regions, we developed DAformer, a novel TL approach that integrates domain adversarial neural networks with the Informer model. Trained on comprehensive runoff data from U.S. basins, DAformer was applied to three basins in Chile and the Chaersen basin in China, demonstrating an effective transfer from data-rich to data-scarce environments. Results show that DAformer significantly outperforms LSTM-based models, improving forecast accuracy by 16.1% for 1-day lead time and by 100.5% for 5-day lead time. These improvements indicate that the DAformer model not only enhances prediction accuracy but also holds substantial practical implications for flood risk management and water resource planning in regions with limited data availability. By clustering basins based on Shuttle Radar Topography Mission (SRTM) and other geographical data, we found that relying on multiple source basins further enhances the performance. DAformer, therefore, serves as a robust and scalable method for enhancing runoff prediction for regions with limited data.

Quantifying the economic value of a national hydrometric network for households

This study reports the results of a Choice Experiment to quantify households’ willingness-to-pay for river gauging programmes in Scotland. The hydrometric network is operated and maintained by the Scottish Environment Protection Agency (SEPA), Scotland’s principal environment regulator, a non-department public body of the Scottish Government. Results from mixed logit and latent class modelling show that most households (‘Hydrometric Maximisers’ − around 70 %) have significant, positive willingness-to-pay values for river gauging programmes, but a minority (‘Hydrometric Satisficers’ − around 30 %) do not view this as a major public policy priority. On average, hydrometric data collection delivers non-market benefits worth £84,625,562 to the Scottish economy, with a minimum economic Benefit-to-Cost ratio of 25:1. This is in addition to the infrastructure value and any private returns made by commercial users of the data. The findings demonstrate that traditional approaches to assessing the benefits of hydrometric networks often underestimate their value. The research also highlights the importance of public information campaigns and household engagement initiatives to increase awareness of hydro-meteorological services, and to develop the business case more fully for public investment in environmental observation networks.

1D-2D hydrodynamic and sediment transport modelling using MIKE models

A comprehensive modeling framework utilizing hydrodynamic and sediment transport models (MIKE Hydro River, MIKE 21 FM, MIKE 21C) was applied to the 33.57 km stretch of the Banshadhara River from Banshadhara bridge, Gunupur to Kashinagar town. Daily discharge, water levels, and sediment data from 2013 to 2022 of Gunupur and Kashinagar gauge stations was used as model input, and calibration–validation of models. Additionally, satellite remote sensing data used for determined Manning’s roughness (n) values as well compared the model outputs. Calibration procedures ensured high accuracy, with MIKE Hydro River achieving 98.3% agreement between observed and simulated water levels. 1D sediment transport model showing significant variability in total bed load values ranging from 0.3 to 3.83 m^3/s. The Wilcock & Crowe gravel bed load formula accurately predicted bed load transport within the expected magnitude, albeit tending to overestimate transport, indicative of supply-limited conditions. 2D MIKE 21 FM simulations highlighted flow dynamics, with rapid flood arrival at Kashinagar and varying current speeds along the river, especially at bends. MIKE 21C, focused on the central portion, indicated erosion processes at river bends with high discharge. Sediment transport model accuracy was approximately 86.7%, capturing general trends despite occasional discrepancies. Over a 10-year simulation, erosion trends and potential river course shifting were observed, particularly near bends. The analysis provides valuable insights for river management, flood risk mitigation, water resource management, infrastructure planning, and environmental preservation in the Banshadhara River basin and similar environments.

Discharge estimation using a rating curve developed for the Upper Erer River, Wabisheble River Basin, Ethiopia

ABSTRACT The measurement of a stream flow is a vital component of water quality monitoring, geomorphology, and flooding investigations. However, the direct measurements of stream flow can be costly and challenging. To obtain a continuous record of discharge, typically the recorded stage and discharge are computed from the correlation of stage and discharge in the form of a rating curve. This study focused on the estimation of flow discharges using a rating curve equation based on data recorded at the Erer-Jijiga Bridge hydrometric station at the outlet of the upper Erer Sub-basin in Ethiopia for 2020–2021. The fitted rating curve was then used to estimate discharges from continuously recorded stages. The installation and operation of a gauging station in the Upper Erer Sub-basin will provide local and regional water sector administrators with more reliable estimates of discharge than would be available if the sub-basin was ungauged. This will also assist the local water sector to undertake water resource planning in the Upper Erer River Basin to better use the limited water resources in the region.

Flood Vulnerability Assessment in Paradip Coastal Area Using Weighted Overlay Index

Paradip, a prominent coastal city on India's eastern seaboard, presents a unique combination of environmental challenges and urban pressures. It has the largest tidal range among major regional tidal gauge stations, including Vishakhapatnam and Chennai. The city experiences an annual sea level rise of 2.33 mm, with projections indicating an increase of 0.183 feet by 2047 and 0.76 feet over the next century. The city's unique environmental and urban issues are exacerbated by its placement within the Mahanadi River's deltaic deposits, floodplains, and coastal marine deposits along the Bay of Bengal. The shoreline of Paradip is topographically uniform, with ground elevations ranging from 3.6 to 5.7 meters. According to India's Wind and Cyclone Hazard Map, Paradip is in the "Very High Damage Risk Zone," indicating the presence of significant wind speeds along the coast. The paper digs into these vulnerabilities and their possible influence on the coastal community, emphasizing the city's need to strike a balance between urban growth and ecological preservation. One of the main problems is flooding. Rainwater enters government buildings and impoverished areas because of encroachments on natural drainage systems. Dryland capable of residential, agricultural, and other economic activities is submerged by overflowing water which causes loss of life, property, and infrastructure. This study evaluates flood vulnerabilities in Paradip, a coastal city, using GIS and a weighted overlay index. By integrating satellite imagery, digital elevation models (DEM), and historical data on rainfall and soil types, we developed a detailed flood vulnerability map that categorises areas into high, moderate, and low risk. ArcGIS Pro 2.7.3 software and a weighted overlay index were utilized for the collection, processing, and display of spatial and attribute data. spatial analysis was the process used to produce the flood vulnerability map. The results provide essential insights for urban planning and flood mitigation strategies, with direct implications for improving disaster preparedness and climate resilience. The flood vulnerability map was then compared to the masterplan of Paradip which was proposed for 2030, to know the actual vulnerable area and to defy some major which were proposed by Paradip development authority. The study illustrates how remote sensing and GIS techniques can be effectively utilized to generate flood vulnerability data essential for informed decision-making.

Leveraging synthetic aperture radar (SAR) with the National Water Model (NWM) to improve above-normal flow prediction in ungauged basins

Abstract Effective flood prediction supports developing proactive risk management strategies, but its application in ungauged basins faces tremendous challenges due to limited/no streamflow record. This study investigates the potential for integrating streamflow derived from synthetic aperture radar (SAR) data and U.S. National Water Model (NWM) reanalysis estimates to develop improved predictions of above-normal flow (ANF) over the coterminous US. Leveraging the SAR data from the Global Flood Detection System to estimate the antecedent conditions using principal component regression, we apply the spatial-temporal hierarchical model (STHM) using NWM outputs for improving ANF prediction. Our evaluation shows promising results with the integrated model, STHM-SAR, significantly improving NWE, especially in 60% of the sites in the coastal region. Spatial and temporal validations underscore the model’s robustness, with SAR data contributing to explained variance by 24% on average. This approach not only improves NWM prediction, but also uniquely combines existing remote sensing data with national-scale predictions, showcasing its potential to improve hydrological modeling, particularly in regions with limited stream gauges.

Methodological evaluation of river discharges derived from remote sensing and land surface models

This study assesses river discharges derived using remote sensing and hydrologic modeling approaches throughout the CONUS. The remote sensing methods rely on total water storage anomalies (TWSA) from the GRACE satellite mission and water surface elevations from altimetry satellites (JASON-2/3, Sentinel-3). Surface and subsurface runoff from two Land Surface Models (NOAH, CLSM) are routed using the Hillslope River Routing model to determine discharge. The LSMs are part of NASA’s Global Land Data Assimilation System (GLDAS). Differences in key physical processes represented in each model, model forcings, and use of data assimilation provide an intriguing basis for comparison. Evaluation is performed using the Kling Gupta Efficiency and USGS stream gauges. Results highlight the effectiveness of both satellite-derived discharge methods, with altimetry generally performing well over a range of discharges and TWSA capturing mean flows. LSM-derived discharge performance varies based on hydroclimatic conditions and drainage areas, with NOAH generally outperforming CLSM. CLSM-derived discharges may be impacted by the use of data assimilation (GLDAS v2.2). Low correlation and high variability contribute to lower KGE values. GLDAS models tend to perform poorly in snow dominated, semi-arid and water-regulated systems where both the timing and magnitude of the simulated results are early and overestimated.

Spatially seamless and temporally continuous assessment on compound flood risk in Hong Kong

Compound flooding results from the simultaneous occurrence of extreme storm surges, sea level rise, and heavy rainfall. These events often lead to impacts significantly more severe than those caused by any individual flood-inducing factor alone. However, the limited and sparse data from tidal gauges hampers precise risk assessment at ungauged sites in coastal cities. Our study addresses this gap by integrating ensemble machine learning with Bayesian inference, offering a comprehensive spatial–temporal analysis of compound flood risk from 1979 to 2022 in Hong Kong. We developed an ensemble machine learning approach within the Bayesian hierarchical modeling framework to achieve spatial–temporal continuity in the estimation of extreme storm surges and mean sea level at sites without tidal gauge stations. Results show a significant yearly increase in maximum storm surge levels by 3 mm and a significant rise in mean sea level of 25 mm per decade in Hong Kong. Our analysis also indicates a significant increase in daily heavy rainfall intensity. Furthermore, in 14.54 % of cases, extreme storm surges coincided with heavy rainfall, while 13.69 % of heavy rainfall events occurred alongside extreme sea level conditions. The copula-based joint analysis reveals significant positive correlations among these extreme events. Our findings further reveal that the return level for a 100-year heavy rainfall event increases dramatically from 126.36 mm in the univariate case to 261.16 mm in the trivariate scenario, underlining the escalated risk associated with compound flooding. Similarly, for storm surge extremes, trivariate analysis reveals elevated risk during compound flood events, with the return level rising from 1.18 m (univariate scenario) to 1.40 m (trivariate scenario) for a 100-year return period. These spatial–temporal maps and comprehensive compound flood risk assessments offer crucial insights for addressing the multi-hazard flood risk in coastal urban areas.

HyG: A hydraulic geometry dataset derived from historical stream gage measurements across the conterminous US

Regional- and continental-scale hydrologic models are increasingly important forecasting tools, yet they rely on highly variable channel parameters (e.g. width, depth, hydraulic resistance) that remain unquantified for millions of stream reaches across the country. Existing hydrologic models utilize over-simplified channel geometries that directly impact the accuracy of streamflow estimates, with cascading effects for water resource and hazard prediction. Here, we present a hydraulic geometry dataset, termed ‘HyG’, derived from discharge field measurements at U.S. Geological Survey (USGS) stream gages across the conterminous United States (CONUS). The HyG dataset includes (1) at-a-station hydraulic geometry parameters, (2) at-a-station hydraulic resistance (Manning’s n) calculated from the Manning equation, (3) daily discharge percentiles, and (4) regionalized downstream hydraulic geometry parameters based on HUC4 (Hydrologic Unit Code 4), derived from a total of 4,051,682 individual measurements from 66,841 total gages. The regionalized HyG dataset can be used directly to improve channel representations in models over CONUS. The original HyG relationships can also be regionalized for finer scales if required.

Utilizing statistical distribution tests to develop rainfall intensity–duration–frequency curves for enhanced hydrological analysis in Kirkuk city, Iraq

ABSTRACT Rainfall intensity is considered one of the basic factors in designing hydrological models based on rainstorm data. The objective of this research is to employ novel intensity–duration–frequency curves and develop empirical equations for rainfall intensity in the city of Kirkuk. The reduction formula adopted by the Indian Meteorological Department was used to divide the maximum daily rainfall for short periods of 0.5, 1, 2, 4, 6, 12, and 24 h. Three key methods of frequency analysis (lognormal distribution, log Pearson type III distribution, and Gumbel extreme value distribution) were utilized to formulate a statistical relationship based on rainfall intensity data from between the years 1981 and 2023 for a gauging station upstream of the Kirkuk city basin to provide the best data set for all periods of 100, 50, 25, 10, 5, and 2 years. Research has shown that the logarithmic distribution is the best fit for modelling the relationship between the annual maximum rainfall at Kirkuk station and duration of the rainfall. The goodness-of-fit results indicate that the lognormal distribution statistically outperforms other distribution models. Hence, the generated rainfall intensity, duration, and frequency curves that were developed led to estimating the intensity of precipitation to build forecasting and hydrological behaviour of the Kirkuk city basin.

Transferability of machine-learning-based modeling frameworks across flood events for hindcasting maximum river water depths in coastal watersheds

Abstract. Despite applications of machine learning (ML) models for predicting floods, their transferability for out-of-sample data has not been explored. This paper developed an ML-based model for hindcasting maximum river water depths during major events in coastal watersheds and evaluated its transferability across other events (out-of-sample). The model considered the spatial distribution of influential factors that explain the underlying physical processes to hindcast maximum river water depths. Our model evaluations in a six-digit hydrologic unity code (HUC6) watershed in the northeastern USA showed that the model satisfactorily hindcasted maximum water depths at 116 stream gauges during a major flood event, Hurricane Ida (R2 of 0.94). The pre-trained, validated model was successfully transferred to three other major flood events, hurricanes Isaias, Sandy, and Irene (R2>0.70). Our results showed that ML-based models can be transferable for hindcasting maximum river water depths across events when informed by the spatial distribution of pertinent features, their interactions, and underlying physical processes in coastal watersheds.

The station system measures the water level and amount of rain that can save the data and online through Web Server

This research paper, the aim is to create a water level and rain gauge station system that can record data and online through Web Server. To be monitor the water level of the river and the amount of water in the rainy season easily by looking through the website. The design of the water level measurement system and the amount of rain is measured with a hydrostatic sensor, Tipping Bucket. A small measuring station, easy to install, uses solar energy and has a Data Logger to record data in parallel with an online database. In the research process, the principle of correlation is used to find the relationship between two variables to find the error of the sensor, namely the water level sensor (Hydrostatic) and the rain sensor (Tipping Bucket). Through research, it was found that the measuring station, the database system can work correctly and with high stability, the water level measurement sensor is highly accurate, the error value is between 0.01 - 0.36 m, the accuracy rate is 97.38%. 98.97% compared to water level measurement stations in nearby areas. For the rain sensor, from the actual count compared to the count with the program 3 times, it is seen that there is 1 error with a deviation value of 0.3 mm. From the recorded data of 2 months (July and August) it was found that the water level is still low to moderate, away from the dangerous water level of 6.84 meters and the amount of rain is still quite high with the amount of rain. The days account for 64.51% and 54.83% of July and August respectively, but the amount of rain will be moderate to moderate. The speed used to send data from the measuring station to the web server via 3G Internet will be between 4 – 6 seconds.

Temporal Variability of Hydroclimatic Extremes: A Case Study of Vhembe, uMgungundlovu, and Lejweleputswa District Municipalities in South Africa

The current study investigated hydroclimatic extremes in Vhembe, Lejweleputswa, and uMgungundlovu District Municipalities based on streamflow data from 21 river gauge stations distributed across the study site for the period spanning 1985–2023. Statistical metrics such as the annual mean and maximum streamflow, as well as trends in annual, maximum, seasonal, and high/low flow, were used to evaluate the historical features of streamflow in each of the three district municipalities. Moreover, the Standardized Streamflow Index (SSI) time series computed from streamflow observations at 3- and 6-month accumulation periods were used to assess hydroclimatic extremes, including drought episodes, proportion of wet/dry years and trends in SSI, drought duration (DD), and drought severity (DS). The results indicate that the three district municipalities have experienced localized and varying degrees of streamflow levels and drought conditions. The uMgungundlovu District Municipality in particular has experienced a significant decline in annual and seasonal streamflow as well as an increase in drought conditions during the 38-year period of analysis. This is supported by the negative trends observed in most of the assessed metrics (e.g., annual, maximum, seasonal, low/high flow, and SSI), whereas DD and DS showed positive trends in all the stations, suggesting an increase in prolonged duration and severity of drought. The Lejweleputswa District Municipality depicted positive trends in most of the assessed metrics, suggesting that streamflow increased, whereas drought decreased in the region over the 38-year period of study. Moreover, the Vhembe District Municipality experienced both negative and positive trends, suggesting localized variations in dry and wet conditions. The results presented in this study contribute towards drought monitoring and management efforts in support of policy- and decision-making that aim to uplift water resources management and planning at local and district municipality levels.