Characterisation of spatial variability in water quality in the Great Barrier Reef catchments using multivariate statistical analysis

Explainable AI approach with original vegetation data classifies spatio-temporal nitrogen in flows from ungauged catchments to the Great Barrier Reef

Transfer of processed data and parameters to ungauged catchments from the most similar gauged counterpart is a common technique in water quality modelling. But catchment similarities for Dissolved Inorganic Nitrogen (DIN) are ill posed, which affects the predictive capability of models reliant on such methods for simulating DIN. Spatial data proxies to classify catchments for most similar DIN responses are a demonstrated solution, yet their applicability to ungauged catchments is unexplored. We adopted a neural network pattern recognition model (ANN-PR) and explainable artificial intelligence approach (SHAP-XAI) to match all ungauged catchments that flow to the Great Barrier Reef to gauged ones based on proxy spatial data. Catchment match suitability was verified using a neural network water quality (ANN-WQ) simulator trained on gauged catchment datasets, tested by simulating DIN for matched catchments in unsupervised learning scenarios. We show that discriminating training data to DIN regime benefits ANN-WQ simulation performance in unsupervised scenarios ( p< 0.05). This phenomenon demonstrates that proxy spatial data is a useful tool to classify catchments with similar DIN regimes. Catchments lacking similarity with gauged ones are identified as priority monitoring areas to gain observed data for all DIN regimes in catchments that flow to the Great Barrier Reef, Australia.

Seasonal estimation of groundwater vulnerability

Index-based methods estimate a fixed value of groundwater vulnerability (GWV); however, the effects of time variations on this estimation have not been comprehensively studied. It is imperative to estimate a time-variant vulnerability that accounts for climatic changes. In this study, we used a Pesticide DRASTICL method separating hydrogeological factors into dynamic and static groups followed by correspondence analysis. The dynamic group is composed of depth and recharge, and the static group is composed of aquifer media, soil media, topography slope, impact of vadose zone, aquifer conductivity and land use. The model results were 42.25-179.89, 33.93-159.81, 34.08-168.74, and 45.56-205.20 for spring, summer, autumn, and winter, respectively. The results showed a moderate correlation between the model predictions and observed nitrogen concentrations with R² = 0.568 and a high correlation for phosphorus concentrations with R² = 0.706. Our results suggest that the time-variant GWV model provides a robust yet flexible method for investigating seasonal changes in GWV. This model is an improvement to the standard index-based methods, making them sensitive to climatic changes and portraying a true vulnerability estimation. Finally, the correction of the rating scale value fixes the problem of overestimation in standard models.

Spatiotemporal analysis and prediction of water quality in Pearl River, China, using multivariate statistical techniques and data-driven model

Identifying spatiotemporal variation patterns and predicting future water quality are critical for rational and effective surface water management. In this study, an exploratory analysis and forecast workflow for water quality in Pearl River, Guangzhou, China, was established based on the 4-h interval dataset selected from 10 stations for water quality monitoring from 2019 to 2021. The multiple statistical techniques, such as cluster analysis (CA), principal component analysis (PCA), correlation analysis (CoA), and redundancy analysis (RDA), as well as data-driven model (i.e., gated recurrent unit (GRU)), were applied for assessing and predicting the water quality in the basin. The investigated sampling stations were classified into 3 categories based on differences in water quality, i.e., low, moderate, and high pollution regions. The average water quality indexes (WQI) values ranged from 38.43 to 92.63. Nitrogen was the most dominant pollutant, with high TN concentrations of 0.81-7.67 mg/L. Surface runoff, atmospheric deposition, and anthropogenic activities were the major contributors affecting the spatiotemporal variations in water quality. The decline in river water quality during the wet season was mainly attributed to increased surface runoff and extensive human activities. Furthermore, the short-term prediction of river water quality was achieved using the GRU model. The result indicated that for both DLCK and DTJ stations, the WQI for the 5-day lead time were predicted with accuracies of 0.82; for the LXH station, the WQI for the 3-day lead time was forecasted with an accuracy of 0.83. The finding of this study will shed a light on an effective reference and systematic support for spatio-seasonal variation and prediction patterns of water quality.

Impact of rainfall characteristics on urban stormwater quality using data mining framework

Understanding the impact of rainfall characteristics on urban stormwater quality is important for stormwater management. Even though significant attempts have been undertaken to study the relationship between rainfall and urban stormwater quality, the knowledge developed may be difficult to apply in commercial stormwater management models. A data mining framework was proposed to study the impacts of rainfall characteristics on stormwater quality. A rainfall type-based calibration approach was developed to improve water quality model performance. Specifically, the relationship between rainfall characteristics and stormwater quality was studied using principal component analysis and correlation analysis. Rainfall events were classified using a K-means clustering method based on the selected rainfall characteristics. A rainfall type-based (RTB) model was independently calibrated for each rainfall type to obtain optimal parameter sets of stormwater quality models. The results revealed that antecedent dry days, average rainfall intensity, and rainfall duration were the most critical rainfall characteristics affecting the event mean concentrations (EMCs) of total suspended solids, total nitrogen, and total phosphorus, while total rainfall was found to be of negligible importance. The K-means method effectively clustered the rainfall events into four types that could represent the rainfall characteristics in the study areas. The rainfall type-based calibration approach can considerably improve water quality model accuracy. Compared to the traditional continuous simulation model, the relative error of the RTB model was reduced by 11.4 % to 16.4 % over the calibration period. The calibrated stormwater quality parameters can be transferred to adjacent catchments with similar characteristics.

Multiple Evaluations of the Spatial and Temporal Characteristics of Surface Water Quality in the Typical Area of the Yangtze River Delta of China Using the Water Quality Index and Multivariate Statistical Analysis: A Case Study in Shengzhou City

Surface water assessments are of critical importance for balancing economic development with the ecological environment in rapidly developing regions. In this research, Shengzhou City, a typical town in the Yangtze River Delta region of China, was chosen to conduct a surface water quality study. As a region with a well-developed water system, monthly water quality monitoring data from eight sampling sites on the major tributaries and the mainstream were selected for six consecutive years from 2013 to 2018, containing seven important water quality indicators (pH, DO, CODMn, CODCr, BOD, NH4+-N, and TP). The comprehensive evaluation method based on the water quality index (WQI) and multivariate statistical analysis methods of cluster analysis (CA) and principal component analysis (PCA) were applied to explore the spatial and temporal changes of water quality in Shengzhou City. The main findings are as follows: (1) spatially, for three main tributaries, Xinchang River had the worst water quality, followed by Changle River, while Huangze River had the best. The water quality of the tributaries had higher volatility than the mainstream. (2) The sampling sites with similar locations had similar water quality characteristics. (3) Seasonally, for the four indicators of DO, CODMn, CODCr, and BOD, the water quality was better in the dry season while, for NH4+-N and TP, water quality was better in the wet season. The low WQI points were more likely to appear in the wet season. (4) The results of WQI assessment showed an improving trend in water quality. (5) Nitrogenous substances and organic matter were the key pollutants in this area. The research results prove that water quality evaluation methods and multivariate statistical methods are effective for the study of regional surface water quality.

Stream water quality optimized prediction based on human activity intensity and landscape metrics with regional heterogeneity in Taihu Basin, China

The driving effects of landscape metrics on water quality have been acknowledged widely, however, the guiding significance of human activity intensity and landscape metrics based on reference conditions for water environment management remains to be explored. Thus, we used the self-organized map, long- and short-term memory (LSTM) algorithm, and geographic detectors to simulate the response of human activity intensity and landscape metrics to water quality in Taihu Lake Basin, China. Fitting results of LSTM displayed that the accuracy was acceptable, and scenario 2 (regional heterogeneity) was more efficient than scenario 1 (regional consistent) in the improvement of water quality. In the driving analysis for the reference conditions, clusters I and II (urban agglomeration areas) were mainly affected by the amount of production wastewater per unit of developed land and the amount of livelihood wastewater per unit of developed land, respectively. Their optimal values were 0.09 × 10³ t/km² (reduction of 35.71%) and 0.2 × 10³ t/km² (reduction of 4.76%). Cluster III (agricultural production areas) was mainly affected by interference intensity, and the optimal value was 2.17 (increased 38.22%), and cluster IV (ecological forest areas) was mainly affected by the fragmentation of cropland, and the optimal value was 1.14 (reduction of 1.72%). The research provides a reference for the prediction of water quality response and presents an ecological and economic sustainability way for watershed governance.

Temporal and spatial variation in water quality in the Yangtze Estuary from 2012 to 2018

Water quality plays an important role in estuarine habitats and affects aquatic organisms. The focus of this study was to understand the temporal-spatial variation of water quality and reveal the natural and anthropogenic drivers by using multiple analysis approaches. Herein, during 2012-2018, six water quality constituents (pH, electrical conductivity (EC), dissolved oxygen (DO), ammonia nitrogen (NH₃N), total nitrogen (TN), and total phosphorus (TP) were examined for eight sites in the Yangtze Estuary, and the hydro-sediment data, i.e., discharge (Q) and sediment transport rate (STR), was collected from the upstream hydrological station Datong. The cluster analysis (CA), principal component analysis (PCA)/factor analysis (FA), Canadian Council of Ministers of the Environment Water Quality Index (CCME-WQI), and the Mann-Kendall (MK) test were applied. The eight sampling sites were geographically grouped into cluster I, cluster II, and cluster III based on the characteristics of water quality changes, which are under the influence of the upstream runoff, the anthropogenic activities, and seawater. The results are as follows: (1) NH₃N, TN, and DO were key constituents representing the water quality of cluster I and cluster III, and those were EC, TN, and DO for cluster II; (2) The monthly-average Q and STR of Datong were negatively correlated to water quality constituents with seasonal variation except for pH; (3) The impact of anthropogenic activities on water quality was especially reflected in cluster III which is close to the Huangpu River Estuary; upstream runoff was the natural driver of water quality changes for cluster I while that was seawater for cluster II. Seawater acts a key role in affecting the water quality of cluster II situated at the North Branch. Revealing the key drivers of water quality change in the Yangtze Estuary provides a reference for water quality management.

Water Treatment Using Natural Coagulant and Electrocoagulation Process: A Comparison Study

Water treatment is the primary consideration before utilizing water for different purposes. Surface water is highly vulnerable to pollution, either due to natural or anthropogenic processes. The main targets of this study were to investigate surface water treatment using Moringa Oleifera (MO), the electrocoagulation process (EC), and the Moringa Oleifera assisted electrocoagulation process (MOAEC). The Moringa Oleifera, EC process, and Moringa Oleifera-assisted EC process are effective mechanisms for the removal of COD (Chemical Oxygen Demand), BOD (Biological Oxygen Demand), TDS (Total Dissolved Solids), phosphate, TSS (Total Suspended Solids), and color from surface water. Different operating parameters such as pH (5–11), the dosage of coagulant (0.2–0.5 g), contact time or reaction time (20–50 minutes), current (0.2–0.5 A), and settling time (5–20 minutes) were considered. The maximum removal efficiency using Moringa Oleifera and the EC process was COD (85.48%), BOD (78.50%), TDS (84.5%), phosphate (95.70%), TSS (93.90%), color (94.50%), and COD (90.50%), BOD (87%), TDS (97.50%), phosphate (89.10%), TSS (95.80%), and color (96.15%), respectively. Similarly, with the application of MOAEC, 91.47%, 89.35%, 97.0%, 90.20%, 9.10%, and 95.70% of COD, BOD, TDS, phosphate, TSS, and color were removed, respectively. The EC process and MOAEC were more effective in the removal of COD, BOD, TDS, TSS, and color than using MO. More phosphate was removed using MO than the EC process and MOAEC. Additionally, the effects of different operating parameters were studied on the removal efficiency.

Classification of catchments for nitrogen using Artificial Neural Network Pattern Recognition and spatial data

In hydrological modelling, classification of catchments is a fundamental task for overcoming deficits in observational datasets. Most attention on this issue has focussed on identifying the catchments with similar hydrological responses for streamflow. Yet, effective methods for catchment classification are currently lacking in respect to Dissolved Inorganic Nitrogen (DIN), a water quality constituent that, at increasing concentrations, is threatening nutrient sensitive environments. Pattern recognition, using standard Artificial Neural Network algorithm is applied, as a novel approach to classify datasets that are considered to be suitable proxies for biological and anthropogenic drivers of observed DIN releases. Eleven gauged Great Barrier Reef (GBR) catchments within Queensland Australia are classified using spatial datasets extracted from ecosystem (e.g. original ecosystem responses to biogeographic, land zone, land form, and soil type attributes) and land use maps. To evaluate the performance of the examined spatial datasets as a proxy for deductive classification, the classification process is repeated inductively, using observed DIN and streamflow data from gauging stations. The ANN-PR method is seen to generate the same classification score format for the differing dataset types, and this facilitates a direct comparison for model output for observed data corroborations. The Kruskal-Wallis test for independence, at p > 0.05, identifies the deductive classification approach as a predictor for classification using DIN observations, which lacks an independence from each other at a p value of 0.01 and 0.02. This study concludes that an ANN-PR method can integrate the ecosystem and land use mapping data to deductively classify the GBR catchments into four regions that also have similar patterns of DIN concentrations. Due to the uniform availability of the mapping data, the findings provide a sound basis for further investigations into the transposing of knowledge from gauged catchments to ungauged areas.

A multi-model approach to assessing the impacts of catchment characteristics on spatial water quality in the Great Barrier Reef catchments

Water quality monitoring programs often collect large amounts of data with limited attention given to the assessment of the dominant drivers of spatial and temporal water quality variations at the catchment scale. This study uses a multi-model approach: a) to identify the influential catchment characteristics affecting spatial variability in water quality; and b) to predict spatial variability in water quality more reliably and robustly. Tropical catchments in the Great Barrier Reef (GBR) area, Australia, were used as a case study. We developed statistical models using 58 catchment characteristics to predict the spatial variability in water quality in 32 GBR catchments. An exhaustive search method coupled with multi-model inference approaches were used to identify important catchment characteristics and predict the spatial variation in water quality across catchments. Bootstrapping and cross-validation approaches were used to assess the uncertainty in identified important factors and robustness of multi-model structure, respectively. The results indicate that water quality variables were generally most influenced by the natural characteristics of catchments (e.g., soil type and annual rainfall), while anthropogenic characteristics (i.e., land use) also showed significant influence on dissolved nutrient species (e.g., NO_X, NH₄ and FRP). The multi-model structures developed in this work were able to predict average event-mean concentration well, with Nash-Sutcliffe coefficient ranging from 0.68 to 0.96. This work provides data-driven evidence for catchment managers, which can help them develop effective water quality management strategies.

Global forest restoration opportunities to foster coral reef conservation

Sediment runoff from disturbed coastal catchments is a major threat to marine ecosystems. Understanding where sediments are produced and where they are delivered enables managers to design more effective strategies for improving water quality. A management strategy is targeted restoration of degraded terrestrial areas, as it provides opportunities to reduce land-based runoff from coastal areas and consequently foster coral reef conservation. To do this strategically, a systematic approach is needed to identify watersheds where restoration actions will provide the highest conservation benefits for coral reefs. Here, we develop a systematic approach for identifying global forest restoration opportunities that would also result in large decreases in the flux of sediments to coral reefs. We estimate how land-use change affects sediment runoff globally using high-resolution spatial data and determine the subsequent risk of sediment exposure on coral reefs using a diffusion-based ocean transport model. Our results reveal that sediment export is a major issue affecting 41% of coral reefs globally. The main coastal watersheds with the highest sediment export are predominantly located in Southeast Asian countries, with Indonesia and the Philippines accounting for 52% of the sediment export in coastal areas near coral reefs. We show how restoring forest across multiple watersheds could help to reduce sediment export to 63,000 km² of coral reefs. Although reforestation opportunities in areas that discharge onto coral reefs are relatively small across watersheds, it is possible to achieve large sediment reduction benefits by strategically targeting watersheds located in regions with a high density of corals near to the coast. Thus, reforestation benefits on coral reefs do not necessarily come from the watersheds that produce the highest sediment export. These analyses are key for generating informed action to support both international conservation policy and national restoration activities.

BoSL FAL pump: A small, low-cost, easily constructed, 3D-printed peristaltic pump for sampling of waters

Water sampling is an essential undertaking for water utilities and agencies to protect and enhance our natural resources. The high variability in water quality, however, often necessitates a spatially distributed sampling program which is impeded by high-cost and large sampling devices. This paper presents the BoSL FAL Pump - a low-cost, easily constructed, 3D-printed peristaltic pump which can be made from commonly available components and is sized to suit even the most space constrained installations. The pump is 38 mm in height and 28 mm in diameter, its components cost $19 AUD and the construction time is just 12 min (excluding 3D printing times). The pump is driven by a direct current motor which is commonly available, cheap and allows for flexibility in the energy supply (5-12 V). Optionally, the pump has a Hall effect sensor and magnet to detect rotation rates and pumping volumes to improve the accuracy of pumping rates/volumes. The pump can be easily controlled by commonly available microcontrollers, as demonstrated by this paper which implements the ATmega328P on the Arduino Uno R3. This paper validates the pump for long-term deployments at flow rates of up to 13 mL per minute in 0.14 mL volume increments at accuracy levels of greater than 99%. The pump itself is scalable, allowing for a wider range of pumping rates when, for example, large volume samples are required for pathogen and micropollutant detection.

Multivariate statistics for spatial and seasonal quality assessment of water in the Doce River basin, Southeastern Brazil

This study employed multivariate statistical techniques in one of the main river basins in Brazil, the Doce River basin, to select and evaluate the most representative parameters of the current water quality aspects, and to group the stations according to the similarity of the selected parameters, for both dry and rainy seasons. Data from 63 qualitative monitoring stations, belonging to the Minas Gerais Water Management Institute network were used, considering 38 parameters for the hydrological year 2017/2018. Principal component analysis (PCA) and hierarchical cluster analysis (HCA) were used to reduce the total number of variables and to group stations with similar characteristics, respectively. Using PCA, four principal components were selected as indicators of water quality, explaining the cumulative variance of 68% in the rainy season and 65% in the dry season. The HCA grouped the stations into four groups in the rainy season and three groups in the dry season, showing the influence of seasonality on the grouping of stations. Moreover, the HCA made it possible to differentiate water quality stations located in the headwaters of the basin, in the main river channel, and near urban centers. The results obtained through multivariate statistics proved to be important in understanding the current water quality situation in the basin and can be used to improve the management of water resources because the collection and analysis of all parameters in all monitoring stations require greater availability of financial resources.

A simulation-based approach to assess the power of trend detection in high- and low-frequency water quality records

To provide more precise understanding of water quality changes, continuous sampling is being used more in surface water quality monitoring networks. However, it remains unclear how much improvement continuous monitoring provides over spot sampling, in identifying water quality changes over time. This study aims (1) to assess our ability to detect trends using water quality data of both high and low frequencies and (2) to assess the value of using high-frequency data as a surrogate to help detect trends in other constituents. Statistical regression models were used to identify temporal trends and then to assess the trend detection power of high-frequency (15 min) and low-frequency (monthly) data for turbidity and electrical conductivity (EC) data collected across Victoria, Australia. In addition, we developed surrogate models to simulate five sediment and nutrients constituents from runoff, turbidity and EC. A simulation-based statistical approach was then used to the compare the power to detect trends between the low- and high-frequency water quality records. Results show that high-frequency sampling shows clear benefits in trend detection power for turbidity, EC, as well as simulated sediment and nutrients, especially over short data periods. For detecting a 1% annual trend with 5 years of data, up to 97% and 94% improvements on the trend detection probability are offered by high-frequency data compared with monthly data, for turbidity and EC, respectively. Our results highlight the benefits of upgrading monitoring networks with wider application of high-frequency sampling.

Tracing the sources of sediment and associated particulate nitrogen from different land uses in the Johnstone River catchment, Wet Tropics, north-eastern Australia

While the ecosystem of the Great Barrier Reef (GBR), north-eastern Australia, is being threatened by the elevated levels of sediments and nutrients discharged from adjacent coastal river systems, the source of these detrimental pollutants are not well understood. Here we used a combined isotopic (δ¹³C, δ¹⁵N) and geochemical (Zn, Pt and S) signatures and stable isotope analysis in R (SIAR) mixing model to estimate the contribution of different land uses to the sediment and associated particulate nitrogen delivered to the Johnstone River. Results showed that rainforest was the largest contributor of suspended and bed sediments in the river estuary (both 33.1%), followed by banana (26.7%, 20.4%), sugarcane (21.5%, 21.4%) and grazing (18.7%, 25.1%). However, bananas and sugarcane land uses had the highest contribution to sediments delivered to the coast per unit of area. This will help land managers to prioritise on-ground activities to improve water quality in the GBR lagoon.

Hydrochemical Characteristics and Water Quality Evaluation of Rivers in Different Regions of Cities: A Case Study of Suzhou City in Northern Anhui Province, China

To study the disparity of river hydrochemical characteristics and water quality in different regions of the city, this paper took the Tuo River in the center of Suzhou, Northern Anhui, China and the Bian River on the edge of the urban area as the research objects, used Piper trigram, Gibbs diagram, and hydrogen and oxygen isotope content characteristics to analyze the geochemical characteristics of surface water in the study area, and then the improved fuzzy comprehensive evaluation method was used to evaluate the water quality. The results showed that the hydrochemical types of the two rivers were SO4-Cl-Na type, and the contents of Na+, K+, SO42−, Cl−, Ca2+, total phosphorus (TP) in the Bian River at the edge of the city were much higher than those in the Tuo River at the center of the city (ANOVA, p < 0.001). Gibbs diagram showed that the ion composition of the two rivers was mainly affected by rock weathering. The results of correlation analysis and water quality evaluation showed that Bian River was greatly affected by agricultural non-point source pollution, and its water quality was poor, class IV and class V water account for 95%, while, for Tuo River, due to the strong artificial protection, class II and class III accounted for 40.74% and 59.26%, respectively, and the overall water quality was better than that of Bian River. The evaluation results of irrigation water quality showed that the samples from Tuo River were high in salt and low in alkali, which could be used for irrigation when the soil leaching conditions were good, while Bian River water samples were high in salt and medium in alkali, which was suitable for irrigation of plants with strong salt tolerance.

Influence of natural and anthropogenic factors on spatial-temporal hydrochemistry and the susceptibility to nutrient enrichment in a subtropical estuary

This study uses multivariate statistics to analyze hydrochemical spatial-temporal variations in the Maowei Sea of Beibu Gulf, South China Sea and evaluates its susceptibility to nutrient enrichment by a risk model. The seasonal variations of sea surface temperature (T), salinity (S), pH, dissolved oxygen (DO), chemical oxygen demand, transparency, total suspended particulate (TSP), petroleum hydrocarbons (PHCs), NO₂^-, and SiO₃^2- were mainly driven by the meteorological factors (precipitation and air temperature), while NO₃^-, NH₄⁺, and PO₄^3- content were more likely related to the point-source factors. The spatial and seasonal variations of T, DO, TSP, PHCs, and SiO₃^2- might also be affected by sea-source factors such as thermal water discharge from adjacent parts of the Beibu Gulf. The sea's susceptibility to nutrient enrichment was moderate, and is mainly affected by precipitation, temperature, and high irradiation. The results present the complexity of natural and anthropogenic influences on a small subtropic estuary.

Supporting Sustainable Development of Water Resources: A Social Welfare Maximization Game Model

Water can carry a boat but can also overturn it (human societal sustainable development). Governments faced aquatic ecosystem restoration and preservation challenges following the establishment of the United Nations Sustainable Development Goals. This paper proposes a social welfare maximization game model to analyze the dominant strategy equilibrium of enterprise-1 and enterprise-2 based on welfare maximization under the total sewage emission control policy. Under the aforementioned control policy, a stricter total sewage emission control of an enterprise corresponds to a lower enterprise output and a higher output of a competing enterprise; that is, the profit transfer effect occurs. When the government implements a relatively strict total sewage emission control policy for an enterprise, it is beneficial to reduce the amount of sewage emission from an enterprise but has no impact on the amount of sewage emission from a competing enterprise; however, the amount of sewage reduction of both enterprises will increase. If the government does not provide capital and technical support to enterprise-2, then enterprise-1 and enterprise-2 should implement total quantity control policies with different rigor degrees to avoid the one-size-fits-all phenomenon. To maximize social welfare, the government should adjust the total sewage emission control policy in time according to sewage stock changes and focus more on enterprises with insufficient capital and poor technical skills and provide financial and technical support.

Understanding the Impacts of Short-Term Climate Variability on Drinking Water Source Quality: Observations From Three Distinct Climatic Regions in Tanzania

Climate change is expected to increase waterborne diseases especially in developing countries. However, we lack understanding of how different types of water sources (both improved and unimproved) are affected by climate change, and thus, where to prioritize future investments and improvements to maximize health outcomes. This is due to limited knowledge of the relationships between source water quality and the observed variability in climate conditions. To address this gap, a 20-month observational study was conducted in Tanzania, aiming to understand how water quality changes at various types of sources due to short-term climate variability. Nine rounds of microbiological water quality sampling were conducted for Escherichia coli and total coliforms, at three study sites within different climatic regions. Each round included approximately 233 samples from water sources and 632 samples from households. To identify relationships between water quality and short-term climate variability, Bayesian hierarchical modeling was adopted, allowing these relationships to vary with source types and sampling regions to account for potentially different physical processes. Across water sources, increases in E. coli/total coliform levels were most closely related to increases in recent heavy rainfall. Our key recommendations to future longitudinal studies are (a) demonstrated value of high sampling frequency and temporal coverage (a minimum of 3 years) especially during wet seasons; (b) utility of the Bayesian hierarchical models to pool data from multiple sites while allowing for variations across space and water sources; and (c) importance of a multidisciplinary team approach with consistent commitment and sharing of knowledge.