This study aimed at establishing a powerful model, using multilayer perceptron (MLP) artificial neural networks, while optimizing database distribution, hidden layer quantity, and node number with the Levenberg–Marquardt learning algorithm, to predict relative humidity in Tangier city, Morocco. The study used a meteorological database with daily readings of 7 variables to predict the output of relative humidity, recorded between January 1985 and December 2022 (13 869 days). The efficacy of the developed models was evaluated by contrasting performance metrics, including coefficient of correlation and mean square error. The best MLP model for the prediction of relative humidity in Tangier has a [7-13-1] architecture, where the hidden layer uses the 'Tansig' function and the output layer uses the 'Purelin' function. The optimized model's efficiency is demonstrated by performance metrics: R = 0.983 and MSE = 0.00295. This MLP model is more efficient than the models established by multiple linear regression, S-MLR and MLR (R ≈ 0.930; MSE ≈ 10.12). These results have practical applications for better understanding and adapting to climate variations in this region.

This study evaluates the utility of Sentinel-2 satellite products for monitoring spatial and temporal changes in chlorophyll a (Chl-a) concentration within an urban, eutrophic estuary. Four atmospheric correction (AC) processors, namely Acolite, C2RCC, Sen2Cor, and Polymer, are assessed together with eight different parameterisations of three high-biomass-appropriate empirical Chl-a retrieval algorithms, namely the 2-band, 3-band and normalised difference chlorophyll index (NDCI). The best performance is achieved using Sen2Cor and the NDCI, which provides an average absolute percentage difference, bias, and correlation of 173%23.8% and 0.853 (p < 0.05), respectively, which improves to 61.7%, 5.4% and 0.843 (p < 0.05), respectively, for conditions where Chl-a > 10 mg/m3. These results indicate that an appropriately configured NDCI algorithm applied to the default Sentinel-2 Level 2 product can be used for routine aquatic water quality monitoring applications for mesotrophic and eutrophic estuaries in the South African context. Monitoring approaches for estuary water quality are essential, as there is an increase in urban runoff and untreated inputs from malfunctioning wastewater treatment systems. The results inform water quality monitoring and management of similar sized estuaries globally. Remote sensing can complement in situ measurements and provide a holistic overview of a system.

Quantification of the water cycle components is key to managing water resources. Remote sensing techniques and products have recently been developed for the estimation of water balance variables. The objective of this study was to test the reliability of LandSAF (Land Surface Analyses Satellite Applications Facility) evapotranspiration (ET) and SPOT-Vegetation Normalised Difference Water Index (NDWI) by comparison with ground-based measurements. Evapotranspiration (both daily and 30 min) was successfully estimated with LandSAF products in a flat area dominated by fynbos vegetation (Riverlands, Western Cape) that was representative of the satellite image pixel at 3 km resolution. Correlation coefficients were 0.85 and 0.91 and linear regressions produced R2 of 0.72 and 0.75 for 30 min and daily ET, respectively. Ground-measurements of soil water content taken with capacitance sensors at 3 depths were related to NDWI obtained from 10-daily maximum value composites of SPOT-Vegetation images at a resolution of 1 km. Multiple regression models showed that NDWI relates well to soil water content after accounting for precipitation (adjusted R2 were 0.71, 0.59 and 0.54 for 10, 40 and 80 cm soil depth, respectively). Changes in NDWI trends in different land covers were detected in 14-year time series using the breaks for additive seasonal and trend (BFAST) methodology. Appropriate usage, awareness of limitations and correct interpretation of remote sensing data can facilitate water management and planning operations.

Historical diatom records provide a means of retrospectively determining water quality and inferring ecological condition in rivers and streams. In this study we re-sampled sites originally sampled 48 years previously. We then determined the scores for the Biological Diatom Index (BDI) and the South African Diatom Index (SADI) for each dataset. The results revealed that the present day conditions in this relatively undisturbed locality were almost identical to those reflected by the samples collected half a century before. This illustrates the value of historical diatom data for the purposes of determining antecedent water quality.

Knowledge of the dynamics of groundwater storage is essential in understanding its driving processes and for informed decision-making. This requires data with adequate precision, which can be obtained from in-situ observations and remote sensing products. While field-based methods are expensive to conduct in low-income countries, satellite products are relatively cheap and provide near real-time data to monitor groundwater resources. This study analyses the seasonal dynamics of groundwater storage anomalies in Zimbabwe. The study utilised the Gravity Recovery and Climate Experiment (GRACE) satellite groundwater monthly point data spanning January 2011 to July 2017. Ordinary kriging was performed to show the spatial variations of groundwater anomaly levels for individual months. Maps for the dry season from April to October were merged together as well as for the wet season from November to March. The final raster maps showed long-term average groundwater level anomalies for each season. Findings show significant variations in groundwater storage level anomalies within and between seasons. The general pattern is that groundwater storage levels increased from the south to north of Zimbabwe. During the wet and dry seasons of 2015, almost 75% of the country had an increase in groundwater storage. This rapid approach using satellite-derived groundwater data, in comparison to other optical remote sensing techniques, is useful for rapid groundwater assessment and management.

A developing trend in stormwater treatment and management is the use of green technologies. Plant biofilters have been gaining increasing use in support of green technology objectives. This technical note reports on the development and preliminary testing of a laboratory-scale plant biofilter prototype for ammonia removal using a South African native plant species (Juncus effusus). The prototype design was based on a conceptual model for nitrogen fixation, plant uptake, bacterial nitrification and soil sorption. Additionally, a plug compartment was incorporated into the design to simulate plug flow as part of the conceptual model. Biofilter models with and without inoculated bacteria were compared. Ammonia reduction, nitrite and nitrate formation were observed. Results showed that the inoculated plant biofilter performed best, with an average of 61% reduction in ammonia within the filter compared to 15% in the normal plant biofilter. The incorporation of a plug compartment aided in slowing down the ammonia infiltration rate, increasing the retention time, and allowing for nitrification to occur.

Irrigation is required to supplement rainfall to enhance the productivity of chickpea in South Africa (SA). However, the dependence on irrigation can be problematic for SA and other countries with limited natural water resources and variable rainfall. Even though access to irrigation water has been identified as one of the challenges faced when planting chickpea in the winter season in SA, irrigation management strategies for chickpea grown on soils differing in texture have not gained considerable research attention. Hence, this study aimed to assess the effects of irrigation levels on dry matter production of chickpea grown on two soils differing in soil texture under greenhouse conditions. The experiment was arranged as a 3 × 2 factorial in a completely randomized design, with 3 irrigation levels (25%, 50% and 75% of the water-holding capacity of soil (WHC)) and 2 soils differing in soil textural class (Loamy sand (LS) soil and sandy loam (SL) soil), replicated thrice. Irrigation level, soil texture and their interaction significantly affected shoot biomass (SBM) and total plant biomass (TBM). Generally, SBM, TBM and root biomass decreased correspondingly with the reduction in irrigation. The 25% WHC significantly reduced the SBM by up to 60% and TBM by up to 56% compared to the 50% and 75% WHC. The SBM and TBM were higher in SL soil than in LS soil. A significantly higher root/shoot ratio (0.45) in the LS soil than in the SL soil (0.16) indicated that the conditions of LS soil encouraged plants to allocate higher proportions of biomass into roots, possibly due to increased competition for soil resources. In conclusion, maintaining soil moisture at 50% WHC ensures better chickpea dry matter production in SL soil.

Higher Himalayan catchments are often poorly monitored for hydrological activities involving flood flow prediction for the safety of riverside communities and the successful operation of hydropower projects. This study aimed to estimate the comparative performance of artificial neural network (ANN) based flow prediction models using 10 years of daily river flow data of Kaligandaki catchment at Kotagaun, Nepal, which is a snow-fed catchment in the Himalayan region. The flow prediction models were trained and tested at a hydrological station using the previous 3 days’ river flow data to predict the 1-day ahead flow data. Eight different training functions were employed in an ANN model for comprehensive statistical assessment of accuracy and precision of each training function. The most significant and validated result obtained in this study is the comprehensive comparison of various training functions’ performance, and identification of the most efficient training function for the study case. Among the training functions investigated, the Levenberg-Marquardt backpropagation function exhibits the best performance for the model having Nash-Sutcliffe efficiency, root mean square error and mean absolute error values of 0.866, 209.578 and 75.422, respectively. This study provides a fundamental basis for accurate flow prediction of topographically challenged catchments where hydrological monitoring and data collection may be limited. In particular, this model will help to improve early warning system, hydrological planning, and the safety of riverside communities in the Himalayan region.

Flow Characteristics (FC) is one of the few methods developed for predicting long-term sustainable borehole yield of single wells in typical fractured rock aquifers. The FC method uses drawdown derivatives and subjective information on no-flow boundaries to estimate a sustainable borehole yield that should not cause the water level to drop below the main water strike (fracture) during long-term operations. Since its development, the FC method has been widely applied in many research and consulting projects. Two decades after its development, a review of its technical capabilities and limitations is necessary to enhance understanding among groundwater practitioners while building a platform for further improvements. The main strength of the method is its simplicity of use, its ability to protect the main water strike/fracture, and its lower susceptibility to the influence of aquifer heterogeneities because it does not require the input of aquifer storativity and transmissivity. The FC method also caters to the negative influence of impermeable boundaries, thereby enabling planning for different low-yield-causing scenarios. However, the major limitation is in using the subjective closed no-flow boundary without factoring aquifer storativity and the distance of the closed no-flow boundary from the pumping well. Under the influence of the closed no-flow boundary, the water must come from aquifer storage, hence the storativity and the size of the bounded aquifer are very critical parameters. It is therefore incorrect to factor in the influence of the closed no-flow boundary without considering its exact location. This limitation is reflected in the absence of criteria to determine the distance of the closed no-flow boundary from the pumping well for validating the FC results using numerical models. The FC method still needs validation using field operational data; other recommendations for future research are highlighted in the discussion.