Community monitoring of coliform pollution in Lake Tanganyika

Happiness Anold Moshi; Daniel Abel Shilla; Ismael Aaron Kimirei; Catherine O’ Reilly; Wim Clymans; Isabel Bishop; Steven Arthur Loiselle

doi:10.1371/journal.pone.0262881

Abstract

Conventional water quality monitoring has been done for decades in Lake Tanganyika, under different national and international programs. However, these projects utilized monitoring approaches, which were temporally limited, labour intensive and costly. This study examines the use of citizen science to monitor the dynamics of coliform concentrations in Lake Tanganyika as a complementary method to statutory and project-focused measurements. Persons in five coastal communities (Kibirizi, Ilagala, Karago, Ujiji and Gombe) were trained and monitored total coliforms, faecal coliforms and turbidity for one year on a monthly basis, in parallel with professional scientists. A standardized and calibrated Secchi tube was used at the same time to determine turbidity. Results indicate that total and faecal coliform concentrations determined by citizen scientists correlated well to those determined by professional scientists. Furthermore, citizen scientist-based turbidity values were shown to provide a potential indicator for high FC and TC concentrations. As a simple tiered approach to identify increased coliform loads, trained local citizen scientists could use low-cost turbidity measurements with follow up sampling and analysis for coliforms, to inform their communities and regulatory bodies of high risk conditions, as well as to validate local mitigation actions. By comparing the spatial and temporal dynamics of coliform concentrations to local conditions of infrastructure, population, precipitation and hydrology in the 15 sites (3 sites per community) over 12 months, potential drivers of coliform pollution in these communities were identified, largely related to precipitation dynamics and the land use.

Citation: Moshi HA, Shilla DA, Kimirei IA, O’ Reilly C, Clymans W, Bishop I, et al. (2022) Community monitoring of coliform pollution in Lake Tanganyika. PLoS ONE 17(1): e0262881. https://doi.org/10.1371/journal.pone.0262881

Editor: Jean-François Humbert, INRA/Sorbonne University, FRANCE

Received: June 7, 2021; Accepted: January 10, 2022; Published: January 28, 2022

This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.

Data Availability: All data are available publicly without restriction: https://doi.org/10.6084/m9.figshare.16610668.

Funding: This study was funded by EU-Horizon 2020 Framework Programme (grant number 776480, MONOCLE and grant number 824603, ACTION). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Since the beginning of 21^st century, many developing countries including Tanzania have experienced rapid urbanization, increasing economic transformation, industrialization and population growth. This has led to an increased pressure on water bodies which receive direct and indirect discharges of wastewater [1], resulting in severe water challenges to local and regional communities [2, 3].

Water quality monitoring is a fundamental tool for the management of lakes, rivers, estuaries, wetlands, and marine systems. With urbanization and agricultural activities impacting waterbodies at an unprecedented level in many African countries, there is a clear need for more information about spatial and temporal dynamics of ecosystem conditions. Such information is fundamental to understand the drivers and identify mitigation strategies necessary to conserve these ecosystems [4, 5].

Microbial and pathogenic pollution can jeopardize public health when lake surface waters are used for drinking, agricultural and recreational uses [6]. A common indicator for assessing water quality for human health is the estimation of coliform bacteria concentrations. However, the determination of coliform concentrations are usually performed by trained scientists in the laboratory [7]. The use of optical approaches can be used as an early indicator of changes in contaminant load, including microbial contamination. Turbidity measurements, for example, are very easy to perform and require minimal training and equipment [8].

Although coliform bacteria are non-pathogenic bacteria that occur in faeces of warm blooded animals, their presence in elevated concentrations has been correlated with various waterborne diseases such as gastrointestinal illness, typhoid fever, ear infections, diarrhoea and dysentery [9]. Presence of coliform bacteria in surface water should at least be seen as a probable threat or indicator of deterioration of microbiological water quality [10] and implies that enteric pathogens may also be present [11]. Coliform bacteria are more numerous, and relatively easy to determine in a laboratory, making them reliable indicators of faecal pollution [12].

Lake Tanganyika, like many African water bodies, is experiencing an increase in pollution loads from local sources. This is particularly true in the small coastal towns in the Kigoma municipality on the north-east shore of the lake, where most lack integrated wastewater treatment systems [12]. Sanitation is limited to pit latrines, pour flush latrines, flushing toilets with septic tanks with a limited incidence of open defecation. Pit latrines are not built for structural strength, but for privacy, and some direct their wastes into Lake Tanganyika. Some members of the fishing community living near the lake defecate directly into Lake Tanganyika offshore [13]. This results in multiple direct health risks to the coastal communities as well as a loss of ecosystem services as local economic activities (eg. fishing and farming) are reliant on lake water and lake waters are used for washing, fishing or drinking [2].

Conventional water quality monitoring has been done for years in Lake Tanganyika under different national and international programs [14–21], that have generated new and important information about lake dynamics and critical tipping points, including the FAO/FINNIDA Lake Tanganyika Research project (1993–1998), the GEF/World Bank Lake Tanganyika Biodiversity Research project (1997–2001), Climate variability in Lake Tanganyika CLIMLake and CLIMFISH (BELSPO) (2002–2006), and the DANIDA Projection of climate change effects in Lake Tanganyika (CLEAT) (2014–2019). However, these projects utilized monitoring approaches that were typically temporally limited, labour intensive and often requiring boats to collect water samples, which were then transported, to the laboratory for analysis. Because such monitoring cannot be easily maintained outside of the scope of these projects, water quality monitoring in Lake Tanganyika has been intermittent and inconsistent. There is a need to explore monitoring techniques that complement existing datasets, can be supported on a long-term basis, and, crucially, can be used locally to inform management actions to improve coastal water quality [22].

A potentially complementary method to provide information on the dynamics of the coastal waters is citizen science [23], a technique that has been used to great extent in monitoring aquatic environments in many parts of the world [24–27]. Citizen science is the partnering of scientists with volunteers to answer scientific questions, reducing costs associated with research projects and creating a more comprehensive data collection [28]. Trained citizen scientists can produce data that is comparable to professionals for a variety of parameters and habitats [29, 30]. Citizen science can also enhance community understanding and awareness of environmental issues, improve conservation efforts, natural resource management and environmental protection [31]. Water quality monitoring has been a particular focus for citizen scientists worldwide, with well-established initiatives like ‘FreshWater Watch’ providing data that can contribute to local, national, and international policy and management decisions [32]. However, most of these initiatives focus on water chemistry and turbidity, which require limited training and relatively inexpensive equipment [33]. Far fewer citizen science studies measure the microbiological contamination due to the cost and complexity of the methods used.

Microbial pollution can have important impacts on local communities in the African Great Lakes [34]. Importantly, microbial pollution events in small coastal communities are often directly related to the activities and local infrastructure that are controlled by the very communities that are impacted. However, without the possibility to monitor these conditions, focused mitigation approaches cannot be applied and validated. In this study, we explore the potential for citizen science to contribute to the monitoring and management of faecal pollution in the freshwater ecosystems that they regularly use for multiple purposes. Specifically, we develop and test a citizen science-based approach for determining changes in coliform concentrations affecting communities on the shores of Lake Tanganyika, Tanzania. Finally, we explore potential drivers related to wastewater management, population growth and land use change and their links to increased pollution periods.

Material and methods

Study design

This study used a longitudinal design with measurements made throughout the year (from May 2019 to April 2020) to explore the spatial and temporal dynamics of coliform and turbidity. Measurements were taken by trained citizen scientists and professional scientists from Tanzania Fisheries Research Institute, Kigoma centre. A total number of 180 water samples were collected and measured by each group of citizen scientists and professional scientists.

Study area

Water samples were collected from lake surface waters in five coastal communities with different levels of environmental and economic pressures. All communities were in the Kigoma municipality and along Lake Tanganyika (Tanzania side): Gombe, Kibirizi, Ujiji, Ilagala and Karago (Fig 1). Three sub-sites were chosen in each community (Table 1). Each measurement site was located between 4 and 1700m away from the lake shores and were chosen based on their different distance from the shore, population size and potential sources of microbiological pollution. Ease of access by professional scientists and citizen scientists was also considered.

Download:

Fig 1. A map of Lake Tanganyika showing study sites.

https://doi.org/10.1371/journal.pone.0262881.g001

Download:

Table 1. Location and description of study sites and associated sub-sites.

https://doi.org/10.1371/journal.pone.0262881.t001

Recruitment and training of citizen science for water quality monitoring

A total of 25 participants, five from each community were recruited from people holding at least Tanzania certificate of secondary education (ordinary level) and older than 18 years of age. Equal opportunity was provided to all qualified participants regardless of gender during the entire recruitment process. Each participant was given a short screening survey to assess their interest in aquatic sciences, environment issues and willingness to participate in lake water quality monitoring. Out of 25 participants, 10 participants (2 in each community) were identified as the potential citizen scientists based on their interest, availability and willingness to participate in lake water quality monitoring (S1 Table). All potential citizen scientists underwent a standard field training and safety course using FreshwaterWatch training materials which include theoretical and hands-on experience on microbial and turbidity measurements. The training was divided into theoretical and practical sessions. For the theoretical session, the training was conducted for two days (16 hours) while the practical session was conducted for 5 days (40 hours). During training, each citizen scientist was required to collect and measure samples of water for total coliform, faecal coliform and turbidity to confirm that they were sufficiently trained and they can participate in the monitoring. Each individual citizen scientist was provided an airtime voucher (for Internet and cell phone services) of 5000 Tanzania shillings (~ €1.85) for each sampling month. Feedback regarding the results and their significance of the measurements was provided to the citizen scientists by the participating researchers. The citizen scientists then transmitted this information to the local communities.

Coliform monitoring

Surface water (0–30 cm) samples for bacterial analysis were collected from each selected sub-site located randomly from 4 to 1700m away from the lake shore (Table 1), following standard sampling guidelines procedures [35]. Both trained citizen scientists and professional scientists from Tanzania Fisheries Research Institute performed identical sample incubation and counting procedures on paired samples over the course of the study. In order to avoid error during sampling, the water samples in each sub-site for citizen scientists and professional scientists were collected in the same bucket and then each group removed the required samples and performed the sample preparation independently on two replicate samples. Each sample was kept in sterilized borosilicate glass stoppered bottles (250 mL), stored in the dark between 6–10°C and transported to the laboratory within 24 hours of sampling.

The membrane filtration method was used for analysis and counting of coliform colony forming units (CFU) per each 100 mL of water sample. Samples were filtered through a 47 mm cellulose nitrate membrane filter, with a 0.45μm pore size. The bacteria trapped on the surface of the filter was then carefully removed and placed in a sterile petri dish containing the solidified media and incubated for 24 hours at 37°C and 44.5°C for total coliforms and faecal coliforms, respectively.

The M-FC Media used for faecal coliform bacteria was prepared by both Tanzania Fisheries Research Institute professional scientists and trained citizen scientists, using Tryptose 10.0 g, Proteose peptone 5.0g, Yeast extract 3.0 g, Sodium Chloride 5.0 g, Lactose 12.5 g, Bile Salts mixture 3 1.5 g, Aniline blue 0.1 g, Agar 15.0 g and 1000 mL of water. The M- ENDO Media used for total coliform bacteria was prepared by both Tanzania fisheries research institute professional scientists and citizen scientists using Tryptone 3.7 g, Peptone 3.7 g, Tryptose 7.5 g, Yeast extract 1.2 g, Lactose 9.4 g, Sodium Chloride 3.7 g, Dipotassium Hydrogen Phosphate 3.30 g, Potassium Dihydrogen Phosphate 1.0 g, Sodium Lauryl Sulphate 0.05 g, Sodium Deoxycholate 0.10 g, Sodium Sulphite 1.6 g, Basic Fuchsin 0.80 g, Agar 15.0 g and 1000 mL of water.

Replicate samples from each site (n = 15) were averaged for each month, providing 180 paired samples (citizen scientist and professional) over one year for comparison and analysis.

Turbidity monitoring

Water samples for turbidity measurement were collected in each sub-site by trained citizen scientists using a clean plastic bottle. After rinsing, a standard and pre-calibrated Secchi tube was used in situ to determine turbidity in nephelometric turbidity units (NTU). Citizen scientists were trained to slowly fill the tube with the water sample until the Secchi disc at the bottom of the tube was no longer visible. The Secchi tube allowed for measurements of turbidity between 14 to 240 NTU [36]. Single measurements were made monthly at each site, allowing for 180 measurements of turbidity of the year.

Rainfall

Lake Tanganyika lies within the east African rift valley and is characterized by a four to five-month cool dry (∼25 °C) season from May to September and a warm (∼28 °C) wet season from October to April. Kigoma receives a mean annual rainfall of about 935 mm and a monthly mean of 36.5 mm [37]. Precipitation data were obtained from Tanzania Meteorological Agency (Kigoma station) which is located approximately 7.2 Km from the lakeshore.

Population and activity drivers

According to the 2012 national census, the regional population is 2,127,930 over an area of 45,066 km², with the highest population densities occurring on the lake shore. The population data for each study site was acquired from the Tanzania National Bureau of Statistics 2012 population and housing census [38].

Land use

In the Kigoma area, natural forest makes up the largest land cover of 45.2%, grazing area of 27.1%, 17.8% waterbodies and 9.9% for other activities, mostly farming. In the present analysis, the dominant land use was assigned as either urban or rural land cover for each site using Google maps and local observations.

Spatial differences in potential drivers related to land use, population and infrastructure in each site were considered, including village population, population density, percentage of households with toilets, distance of the site from households, level of economic activity, farming intensity, number of livestock, slope of the land with respect to the lake, closed or open bays and urban/rural dominance. These physical and socio-economic characteristics were estimated from available population and housing census data, with distance from households being that from the site to the nearest house. Binary values (1 and 0) were assigned by the authors for non-continuous data (economic activity, farming intensity, closed/open bays, urban/rural dominance) based on available data and personal observations. A value of 1 was given for high economic activity (active market or small industry) and 0 was assigned for low economic activity. A value of 1 was also given to high farming intensity (cultivation area greater than 30% of total area) and 0 for low farming intensity (cultivation area less than 30% of the total area [39]). Closed/open bays, urban/rural dominance were based on visual comparisons using Google Maps. Ordinal values were given to continuous values such as rural/urban dominance, closed/open bay (Table 2).

Download:

Table 2. Characteristics of the five study sites.

https://doi.org/10.1371/journal.pone.0262881.t002

Statistical analysis

Data were analysed using standard methods, including paired t-Test (two-paired sample for means), one-way Analysis of Variance (ANOVA), Pearson and Spearman correlations. Multiple regression was performed to understand the significance of potential factors influencing coliform concentration, including: population, population density, farming intensity, economic activity, slope of the land to the lake, households with toilets, number of livestock and precipitation with total and faecal coliform concentration as dependent factors. All data were tested with an alpha level of significance of 0.05 and using a Bonferroni correction for multiple correlations using Realstats (version 2016). Measurements and data collection were not impacted by Covid-19 restrictions.

Results

Temporal dynamics of coliform concentrations and turbidity

Total coliform concentrations showed a strong seasonal dynamic, with higher concentrations measured in the rainy season (905 CFU/100 mL) compared to the dry season (171 CFU/100 mL) (t = 9.5, p < 0.001). A similar dynamic was observed for faecal coliforms, with a lower t-statistic, but following the same trend with higher rainy season concentrations (288 CFU/100 mL) compared to the dry season (114 CFU/100 mL) (t = 4.6, p < 0.001). Study sites differed in their seasonal coliform dynamics (Fig 2A and 2B). Turbidity also showed a seasonal behaviour, with the rainy season having a higher mean (32 NTU) with respect to the dry season (17 NTU) (t = 3.0, p = 0.003) (Fig 2C).

Download:

Fig 2. Monthly mean concentrations (in line) ± standard deviation (shaded area) of coliforms and turbidity determined by citizen scientists in each study site.

(a) Mean total coliform counts (CFU/100 mL) ± standard deviation, (b) Mean fecal coliform count (CFU/100 mL) ± standard deviation (c) Mean turbidity (NTU) ± standard deviation.

https://doi.org/10.1371/journal.pone.0262881.g002

Spatial Dynamics of Coliform and TurbidityThere was a clear difference in the average total coliform concentration between sampling sites (ANOVA, df = 4, F = 5.2, p <0.001; Table 3). The TC counts were highest at Ilagala and lowest at Gombe and Karago, with an overall average of 568±246 CFU/100mL, considering all sampling sites. Likewise, there was a significant difference in faecal coliform among sites (df = 4, F = 6.7, p <0.001; Fig 3B). Ujiji displayed the highest FC counts with the lowest at Gombe and Karago. The overall average of all sampling sites was 188±88 CFU/100mL. Similarly, turbidity concentrations were significantly different among site (df = 4, F = 5.4, p = 0.003; Fig 3C), with the highest turbidity at Ilagala and the lowest again at Gombe.

Download:

Fig 3. Boxplot representing mean concentrations ± standard deviation of coliform and turbidity determined by citizen scientists in 5 study sites (n = 36 per site).

(a) Mean total coliform counts (CFU/100 mL) (b) Mean fecal coliform counts (CFU/100mL) (c) Mean Turbidity (NTU). Bars represent standard deviation and points represent variation around mean.

https://doi.org/10.1371/journal.pone.0262881.g003

Download:

Table 3. Average and standard deviation of coliform and turbidity for each study site (n = 36).

https://doi.org/10.1371/journal.pone.0262881.t003

Potential drivers for coliform variabilities

Multiple linear regression showed low effect size for total coliforms and faecal coliforms (r² = 0.31 and 0.17, p < 0.001), with the only significant coefficient being that related to monthly average precipitation. The regression with turbidity (r² = 0.11, p = 0.02) had significant coefficients for monthly average precipitation as well as the relative distance of the site from households.

By comparing concentrations with respect to local conditions, economic activity level, distance of the site from household, farming intensity, river impact and urban or rural dominance, it was possible to identify which site attributes were linked to higher coliform or turbidity conditions. Significant differences (T-test, p < 0.05, Table 4) in average total coliform concentrations were found for sites with: higher economic activity, shorter distance from households, higher farming intensity, river proximity and urban dominance. Significant difference in average faecal coliform concentrations were found for sites with: higher economic activity, higher farming intensity, river proximity and urban dominance. Higher turbidity was associated with sites with higher economic activity, shorter distance from households, higher farming intensity, river proximity and urban dominance.

Download:

Table 4. t-Test: Paired two sample for means showing the monthly mean ± Standard Deviation (S.D) of potential drivers influencing concentration of coliform and turbidity in the study sites separated by different characteristic, t-static and p-value at 0.05.

The variables which were significant are bolded.

https://doi.org/10.1371/journal.pone.0262881.t004

Seasonal precipitation dynamics were seen as a potential driver of coliform and turbidity dynamics. We compared monthly precipitation (total mm/month) to coliform and turbidity averages for each study site (Pearson correlation, Table 5). Faecal coliforms showed the highest correlation with precipitation in Ilagala and Kibirizi, while the correlations were much poorer for the other sites, indicating that the dynamics of these latter sites were not well associated to those of monthly precipitation. Total coliform dynamics showed a better correlation with precipitation, with significant correlations with monthly precipitation dynamics in all sites except Ujiji. Monthly dynamics for turbidity were most similar to monthly precipitation in Ujiji and Ilagala.

Download:

Table 5. Comparison of t-scores related to Pearson correlations comparing monthly dynamics of coliform and turbidity with monthly dynamics of precipitation for each study site.

https://doi.org/10.1371/journal.pone.0262881.t005

Relationship between turbidity and coliforms

While the temporal dynamics of monthly averaged coliform and turbidity followed a similar trend (Fig 2), there was no significance correlation between these individual measurements (r < 0.10, p > 0.05). However, when the coliform data were divided into measurements taken in conditions of high turbidity (> 14 NTU) and low turbidity (≤ 14 NTU), clear differences emerged. The mean and median concentrations of faecal coliforms were significantly different for the two turbidity categories (Fig 4) (T-test, p < 0.001). For total coliforms, there was no significant difference.

Download:

Fig 4. Boxplots of Mean coliform counts (CFU/100mL) determined by citizen scientists based on turbidity measurements, above and below 14 NTU cutoff points.

FC_NTU<14 = Mean fecal coliform counts below 14 NTU, FC_NTU<14 = Mean fecal coliform counts above 14 NTU, TC_NTU<14 = Mean total coliform counts below 14 NTU and TC_NTU>14 = Mean total coliform counts above 14 NTU. Bars represent standard deviation and points represent variation around mean.

https://doi.org/10.1371/journal.pone.0262881.g004

Considering a turbidity cut-off of 14 NTU, it was possible to evaluate whether Secchi tube measurements could be a proxy for local conditions of high coliform concentrations. The 14 NTU cut-off was associated to a depth of 433 mm or 95% of the total length of a calibrated Secchi tube. When turbidity is below 14 NTU, there was a 76% probability of faecal coliform concentration to be below 200 CFU/100mL. For the same turbidity cut-off, there was a 70% probability that total coliform counts were below 600 CFU/100mL. Turbidity along the coastal area of Lake Tanganyika can be highly variable and is strongly influenced by the presence of rivers and coastal activities [51], with open lake values ranging from 2 to 21 NTU [52].

Comparison of citizen and professional scientist measurements

The correlation for total coliforms between professional scientists and citizen scientists was very strong (Spearman rho = 0.99, n = 180) (Fig 5A), while the regression equation showed a minor overestimation of citizen scientists’ values over professional scientists’ values; (professional scientists CFU = 0.96*citizen scientists CFU– 17.5 CFU, p <0.001).

Download:

Fig 5.

Pearson correlation of total coliform and fecal coliform concentration (CFU/100 mL) observed by professional and citizen scientists (a) Total coliform (CFU/100 mL) (b) fecal coliform (CFU /100m/L) (n = 180).

https://doi.org/10.1371/journal.pone.0262881.g005

The correlation for faecal coliforms between professional scientists and citizen scientists was also very strong (Spearman rho = 0.94, n = 180) (Fig 5B), while the regression equation showed a slightly larger overestimation of citizen scientists’ values with respect to professional scientists values; (professional scientists CFU = 0.90*citizen scientists CFU– 0.44 CFU, p <0.001).

Discussion

Faecal contamination of near shore waters in many African lakes is a major challenge for local communities and national public health agencies, as people in the coastal communities often use lake water for cooking and washing. The concentrations observed in the present study are similar to those measured in Lake Malawi, with coliform concentrations of 35-400CFU/100mL in nearshore surface waters [40], and lower than those reported for Lake Victoria’s fish landing sites (3.6×10⁶ CFU/mL, [41] and Lake Victoria’s Nakivubo channel (up to 9.2×10⁶ CFU/mL, [42]. In the Burundian surface waters of Lake Tanganyika, concentrations are similarly elevated, averaging 33250CFU/100mL and 2000CFU/100mL for total and faecal coliforms, respectively [1]; similar to our study, population density seems to be a major driver.

In the present study, Ilagala, Kibirizi and Ujiji presented the highest coliform concentrations, as well as a high sensitivity to precipitation. Ilagala has a high population and receives effluents from Malagarasi river, the second largest river in Tanzania. The effluent load becomes particularly elevated in the rainy season when the river can overflow. Kibirizi has an important fishery industry, with the largest fish market among the studied sites and related number of visitors. Kibirizi is also located in a closed bay, with a higher retention of runoff. While not receiving any discharge from any major river, Kibirizi receives waters from temporal streams which flow into the lake during the rainy season. These streams drain surrounding areas which have small population centres as well as agricultural activities.

Gombe and Karago presented the best conditions, with the lowest coliform concentrations and turbidity of the study sites. Gombe is a small community, located near a national park where most economic activities are prohibited. Most of the area around Gombe is mountainous and well vegetated, which reduces possible influx of livestock and human wastes. Karago is a small rural area in a semi-closed bay, with a low population and animal density. Karago showed no significant linkage between faecal coliform and precipitation dynamics, suggesting that large quantities of wastewater do not flow directly into the lake during heavy rain events. Both sites had low economic and agricultural activities, leading to a low use of animal wastes and other pollutants that might increase contamination and particle concentrations in the nearby lake waters.

Ujiji presents mixed conditions, with high coliforms and turbidity, but a low influence from precipitation. In fact, Ujiji has the lowest relative monthly variance of total and faecal coliforms of any of the study sites, indicating a more regular emission and a lower sensitivity to seasonal patterns in climate. Ujiji has a population similar to that Ilagala and is also located near the discharge of river. The Luiche River drains in Ujiji area and several agricultural villages. River borne particulate and dissolved matter from agricultural activities, grazing and other anthropogenic activities can be significant [43, 44]. The combination of close proximity of the houses and the influence of the Luiche River both lead to conditions of high coliform concentrations and high turbidity in Ujiji, with limited seasonal variation.

The comparative study of coliform concentrations measured by trained and equipped citizen scientists to those determined by professional scientists show a very high correlation, with a small overestimation for both total and faecal coliforms. This suggests that citizen scientists, well trained and equipped, can make accurate estimates of coliform concentrations that could be used to complement municipal and national monitoring efforts. Sampling in periods of high risk (eg. rainy season) or local community citizen scientists in collaboration with laboratories in Kigoma could perform suspected moments of high coliforms. Interestingly, measurements of high turbidity provided information on conditions of elevated coliform loads, allowing for the creation of a tiered approach to monitoring coastal lake conditions and informing local communities of high risk conditions.

The use of turbidity to estimate loads of coliform bacteria is attractive due to the rapidity with which turbidity can be determined automatically [45] through in situ sensors [46] or by drone [47] or satellite sensing [48, 49]. In the present study, manual measurements of turbidity were made by trained citizen scientists using a simple calibrated Secchi tube. A turbidity cut-off of 14 NTU provided good accuracy for separating periods of low and high faecal and total coliform concentrations. This implies that when turbidity is higher than 14 NTU, there is an elevated possibility of coliform pollution, which warrants further testing or possibly, community-based restrictions on lake water use until conditions improve.

During and after rain events, surface runoff increases the possibility that dissolved and particulate matter from agricultural areas and human settlements are carried into the lake. Studies have shown that coliform dynamics in surface waters can be strongly modified by local precipitation [50–52]. In Lake Victoria, higher concentrations of microbial pollutants in the wet season have been associated to storm water run-offs from multiple sources [53] while studies in different climate zones have found similar results [54–56]. It should be noted that coastal conditions can be highly variable after rain events, showing the importance of replicate sampling during the rainy season [7, 57]. Our study was limited to regularly monthly sampling and not focused specifically on the rain events.

The concentration of coliform across the sampling sites was also found to be influenced by farming intensity, number of toilets, population, number of animals and rural versus urban dominated land uses (Table 2). Anthropogenic factors, such as land use and wastewater treatment have often been associated to elevated coliform concentrations [56, 58], as have agricultural activities requiring the use or storage of animal manure [59]. Further study would be required to identify the specific links between different land use types and coliform contamination in lake waters to develop appropriate mitigation measures [60]. This kind of information could be gained from the use of similar citizen science methods across more coastal communities of Lake Tanganyika.

Conclusions

According to Tanzania guidelines for drinking water [61], the permissible faecal coliform for drinking water is less than 1 CFU/100 mL. There were no measurements (n = 180) taken over the 12- months period where this standard was met. A more appropriate standard is the WHO guideline for permissible faecal coliform counts for recreational purposes of 200 CFU/100 mL or that for irrigation purposes of 1000 CFU/100 mL [62]. Using the former, community citizen scientists along the shores of Lake Tanganyika could use calibrated standard Secchi tubes to identify conditions where lake water is no longer appropriate for direct use (eg. washing) without further treatment. Using a tiered approach, a first screening result of high turbidity could be followed by further testing for coliforms, reducing the costs and challenges related to the training and equipment necessary for coliform sampling and identification. Developing a sampling strategy based on lower cost approaches can represent a powerful tool to monitor changes in water quality when applied at high frequency [63]. There is a clear opportunity for local populations to obtain high frequency data using a $2 Secchi tube when regulatory monitoring by professional technicians is limited [64]. With developments in simple optical tools for smartphones, there is a growing opportunity for rapid assessment of lake conditions by community members. Citizen science programmes have costs related to recruiting, training and equiping volunteers, creating an open database that can receive monitoring data and perform associated quality control, as well as feedback and exchange with active participants. These costs will depend on the number of participants and availability of local scientists.

These results suggest that communities can contribute to monitoring their local waters for microbiological conditions. Similar successes by citizen scientists monitoring benthic communities [65], phytoplankton concentrations [66] and nutrient pollution [30] have been reported. Given that the infrastructure for coastal communities is unlikely to change over the short term, the results of the present study indicate that community-based monitoring, with trained local citizen scientists, could inform municipal and national agencies as well as those same communities of poor water quality conditions as well as validate the efficacy of mitigation actions and management strategies initiated by the same community.

Supporting information

S1 Table. Demographic characteristics of citizen scientists participated in coliform and turbidity monitoring in Lake Tanganyika.

https://doi.org/10.1371/journal.pone.0262881.s001

(DOCX)

S1 Graphical abstract.

https://doi.org/10.1371/journal.pone.0262881.s002

(TIF)

Acknowledgments

The authors would like to thank the hard work of the citizen scientists from Kigoma communities involved in this research. We are grateful for the support and collaboration of colleagues at Tanzania Fisheries Research Institute (TAFIRI) Kigoma centre during sampling and laboratory analysis, Earthwatch Europe for training and support of citizen science materials. We express our sincere thanks to Kigoma zonal water quality laboratory for their assistance during laboratory sample analysis.

References

1. Niyoyitungiye L, Giri A, Ndayisenga M. Assessment of Coliforms Bacteria Contamination in Lake Tanganyika as Bioindicators of Recreational and Drinking Water Quality. South Asian Journal of Research in Microbiology. 2020:9–16.
- View Article
- Google Scholar
2. Chen Kimirei IA. Demonstration Research on Comprehensive Water Quality Monitoring in the Lake Tanganyika Basin. CSTEC and NIGLAS, Nanjing, China. 2015.
3. Shen Q, Zhang L, Kimirei IA, Wang Z, Gao Q, Chen S, et al. Vertical physicochemical parameter distributions and health risk assessment for trace metals in water columns in eastern Lake Tanganyika, Tanzania. Journal of Oceanology and Limnology. 2019;37(1):134–145.
- View Article
- Google Scholar
4. Stutter M, Langan S, Cooper R. Spatial and temporal dynamics of stream water particulate and dissolved N, P and C forms along a catchment transect, NE Scotland. Journal of Hydrology. 2008;350(3–4):187–202.
- View Article
- Google Scholar
5. Zhang Y, Ma R, Hu M, Luo J, Li J, Liang Q. Combining citizen science and land use data to identify drivers of eutrophication in the Huangpu River system. Science of the Total Environment. 2017;584:651–664. pmid:28132775
- View Article
- PubMed/NCBI
- Google Scholar
6. Cho S, Hiott LM, Barrett JB, McMillan EA, House SL, Humayoun SB, et al. Prevalence and characterization of Escherichia coli isolated from the Upper Oconee Watershed in Northeast Georgia. PloS One. 2018;13(5):e0197005. pmid:29738574
- View Article
- PubMed/NCBI
- Google Scholar
7. Tornevi A, Bergstedt O, Forsberg B. Precipitation effects on microbial pollution in a river: lag structures and seasonal effect modification. PloS One. 2014;9(5):e98546. pmid:24874010
- View Article
- PubMed/NCBI
- Google Scholar
8. Miguel-Chinchilla L, Heasley E, Loiselle S, Thornhill I. Local and landscape influences on turbidity in urban streams: a global approach using citizen scientists. Freshwater Science. 2019;38(2):303–320.
- View Article
- Google Scholar
9. Gruber JS, Ercumen A, Colford JM Jr. Coliform bacteria as indicators of diarrheal risk in household drinking water: systematic review and meta-analysis. PloS One. 2014;9(9):e107429. pmid:25250662
- View Article
- PubMed/NCBI
- Google Scholar
10. Ndlovu T, Le Roux M, Khan W, Khan S. Co-detection of virulent Escherichia coli genes in surface water sources. PloS One. 2015;10(2):e0116808. pmid:25659126
- View Article
- PubMed/NCBI
- Google Scholar
11. Perkins TL, Clements K, Baas JH, Jago CF, Jones DL, Malham SK, et al. Sediment composition influences spatial variation in the abundance of human pathogen indicator bacteria within an estuarine environment. PloS One. 2014;9(11):e112951. pmid:25397595
- View Article
- PubMed/NCBI
- Google Scholar
12. Gerba CP. Indicator microorganisms. Environmental Microbiology: Elsevier; 2009. p. 485–499.
13. Nkotagu HH. Lake Tanganyika ecosystem management strategies. Aquatic Ecosystem Health & Management. 2008;11(1):36–41.
- View Article
- Google Scholar
14. Azanga OE. Land-use and land cover, sediment and nutrient hotspot areas changes in Lake Tanganyika Basin. African Journal of Rural Development. 2016;1(1):75.
- View Article
- Google Scholar
15. Bergamino N, Horion S, Stenuite S, Cornet Y, Loiselle S, Plisnier P-D, et al. Spatio-temporal dynamics of phytoplankton and primary production in Lake Tanganyika using a MODIS based bio-optical time series. Remote Sensing of Environment. 2010;114(4):772–780.
- View Article
- Google Scholar
16. Cohen AS, Palacios-Fest MR, Msaky ES, Alin SR, McKee B, O’Reilly CM, et al. Paleolimnological investigations of anthropogenic environmental change in Lake Tanganyika: IX. Summary of paleorecords of environmental change and catchment deforestation at Lake Tanganyika and impacts on the Lake Tanganyika ecosystem. Journal of Paleolimnology. 2005;34(1):125–145.
- View Article
- Google Scholar
17. Donohue I, Irvine K. Seasonal patterns of sediment loading and benthic invertebrate community dynamics in Lake Tanganyika, Africa. Freshwater Biology. 2004;49(3):320–331.
- View Article
- Google Scholar
18. Kalacska M, Arroyo-Mora JP, Lucanus O, Kishe-Machumu MA. Land cover, land use, and climate change impacts on Endemic Cichlid Habitats in Northern Tanzania. Remote Sensing. 2017;9(6):623.
- View Article
- Google Scholar
19. Karamage F, Shao H, Chen X, Ndayisaba F, Nahayo L, Kayiranga A, et al. Deforestation effects on soil erosion in the Lake Kivu Basin, DR Congo-Rwanda. Forests. 2016;7(11):281.
- View Article
- Google Scholar
20. Patterson G, Makin J. The state of biodiversity in Lake Tanganyika; a literature review. Chatham, UK: Natural Resource Institute; 1998. 1998.
21. Mziray P, Kimirei IA, Staehr PA, Lugomela CV, Perry WL, Trolle D, et al. Seasonal patterns of thermal stratification and primary production in the northern parts of Lake Tanganyika. Journal of Great Lakes Research. 2018;44(6):1209–1220.
- View Article
- Google Scholar
22. Kelly B, Mtiti E, McIntyre PB, Vadeboncoeur Y. Stable isotopes reveal nitrogen loading to Lake Tanganyika from remote shoreline villages. Environmental Management. 2017;59(2):264–273. pmid:27826695
- View Article
- PubMed/NCBI
- Google Scholar
23. Dickinson JL, Shirk J, Bonter D, Bonney R, Crain RL, Martin J, et al. The current state of citizen science as a tool for ecological research and public engagement. Frontiers in Ecology and the Environment. 2012;10(6):291–297.
- View Article
- Google Scholar
24. Hughes RN, Hughes DJ, Smith IP. Citizen scientists and marine research: volunteer participants, their contributions, and projection for the future. Oceanography and Marine Biology: An Annual Review, Volume 52. 2014;52:257–314.
- View Article
- Google Scholar
25. Hyder K, Wright S, Kirby M, Brant J. The role of citizen science in monitoring small-scale pollution events. Marine Pollution Bulletin. 2017;120(1–2):51–57. pmid:28477985
- View Article
- PubMed/NCBI
- Google Scholar
26. Loiselle A S, Gasparini Fernandes Cunha D, Shupe S, Valiente E, Rocha L, et al. Micro and macroscale drivers of nutrient concentrations in urban streams in South, Central and North America. PloS One. 2016;11(9):e0162684. pmid:27662192
- View Article
- PubMed/NCBI
- Google Scholar
27. Thornhill I, Loiselle S, Clymans W, van Noordwijk C. How citizen scientists can enrich freshwater science as contributors, collaborators, and co-creators. Freshwater Science. 2019;38(2):231–235.
- View Article
- Google Scholar
28. Ceccaroni L, Bowser A, Brenton P. Civic education and citizen science: Definitions, categories, knowledge representation. Analyzing the role of citizen science in modern research: IGI Global; 2017. p. 1–23.
29. Gillett DJ, Pondella DJ, Freiwald J, Schiff KC, Caselle JE, Shuman C, et al. Comparing volunteer and professionally collected monitoring data from the rocky subtidal reefs of Southern California, USA. Environmental Monitoring and Assessment. 2012;184(5):3239–3257. pmid:21739280
- View Article
- PubMed/NCBI
- Google Scholar
30. McGoff E, Dunn F, Cachazo LM, Williams P, Biggs J, Nicolet P, et al. Finding clean water habitats in urban landscapes: professional researcher vs citizen science approaches. Science of the Total Environment. 2017;581:105–116. pmid:28069307
- View Article
- PubMed/NCBI
- Google Scholar
31. McKinley DC, Miller-Rushing AJ, Ballard HL, Bonney R, Brown H, Cook-Patton SC, et al. Citizen science can improve conservation science, natural resource management, and environmental protection. Biological Conservation. 2017;208:15–28.
- View Article
- Google Scholar
32. Quinlivan L, Chapman DV, Sullivan T. Validating citizen science monitoring of ambient water quality for the United Nations sustainable development goals. Science of the Total Environment. 2020;699:134255. pmid:31683215
- View Article
- PubMed/NCBI
- Google Scholar
33. Buytaert W, Zulkafli Z, Grainger S, Acosta L, Alemie TC, Bastiaensen J, et al. Citizen science in hydrology and water resources: opportunities for knowledge generation, ecosystem service management, and sustainable development. Frontiers in Earth Science. 2014;2:26.
- View Article
- Google Scholar
34. Fuhrimann S, Stalder M, Winkler MS, Niwagaba CB, Babu M, Masaba G, et al. Microbial and chemical contamination of water, sediment and soil in the Nakivubo wetland area in Kampala, Uganda. Environmental Monitoring and Assessment. 2015;187(7):475. pmid:26122126
- View Article
- PubMed/NCBI
- Google Scholar
35. APHA. Standard methods for the examination of water and wastewater, American Public Health Association, American Water Works association, Water Environment Federation. 1998.
36. White T. Monitoring a watershed: Nationwide turbidity testing in Australia. Volunt Monit. 1994;6:22–23.
- View Article
- Google Scholar
37. Hunink J, Terink W, Contreras S, Droogers P. Scoping Assessment of erosion levels for the Mahale region, Lake Tanganyika, Tanzania. Unpubl FutureW Rep. 2015;148:1–47.
- View Article
- Google Scholar
38. NBS. Tanzania National Bureau of statistics; 2012 population and housing census. 2013.
39. URT. Kigoma Socio-economic Profile, Ministry of Planning, Economy and Empowerment, Dar es Salaam. 2016.
40. Samikwa E, Kapute F, Tembo M, Phiri T, Holm RH. Identification of critical control points using water quality as an indicator of hygiene for artisanal fisheries on Lake Malawi. Lakes and Reservors: Research and Management. 2019;24(1):3–12.
- View Article
- Google Scholar
41. Odongkara K, and Kyangwa I. Sanitation, fish handling and artisanal fish processing within fishing communities: socio-cultural influences.Jinja, Uganda,National Fisheries Resources Research Institute (NaFIRRI). 2005:46pp.
42. Byamukama D, Kansiime F, Mach RL, Farnleitner AH. Determination of Escherichia coli contamination with chromocult coliform agar showed a high level of discrimination efficiency for differing fecal pollution levels in tropical waters of Kampala, Uganda. Applied Environmental Microbiology. 2000;66(2):864. pmid:10653767
- View Article
- PubMed/NCBI
- Google Scholar
43. Langenberg VT, Nyamushahu S, Roijackers R, Koelmans AA. External nutrient sources for Lake Tanganyika. Journal of Great Lakes Research. 2003;29:169–180.
- View Article
- Google Scholar
44. Nkotagu HH, Athuman C. The limnology of the Lake Tanganyika sub catchment. Tanzania Journal of Science. 2007;33.
- View Article
- Google Scholar
45. Christensen VG, Rasmussen PP, Ziegler AC, Jian X, editors. Continuous monitoring and regression analysis to estimate bacteria loads. Water Environment Federation, TMDL Science Issues Conference, St Louis, Missouri; 2001.
46. Loperfido J, Just CL, Papanicolaou AN, Schnoor JL. In situ sensing to understand diel turbidity cycles, suspended solids, and nutrient transport in Clear Creek, Iowa. Water Resources Research. 2010;46(6).
- View Article
- Google Scholar
47. De Keukelaere L, Sterckx S, Adriaensen S, Bhatia N, Monbaliu J, Toorman E, et al. Coastal turbidity derived from PROBA-V global vegetation satellite. Remote Sensing. 2020;12(3):463.
- View Article
- Google Scholar
48. Liu G, Wei J, Muthu B, Jackson Samuel RD. Chlorophyll-a concentration in the hailing bay using remote sensing assisted sparse statistical modelling. European Journal of Remote Sensing. 2020:1–12.
- View Article
- Google Scholar
49. Tyler AN, Hunter PD, Spyrakos E, Groom S, Constantinescu AM, Kitchen J. Developments in Earth observation for the assessment and monitoring of inland, transitional, coastal and shelf-sea waters. Science of the Total Environment. 2016;572:1307–1321. pmid:26805447
- View Article
- PubMed/NCBI
- Google Scholar
50. Bastaraud A, Perthame E, Rakotondramanga J-M, Mahazosaotra J, Ravaonindrina N, Jambou R. The impact of rainfall on drinking water quality in Antananarivo, Madagascar. PloS One. 2020;15(6):e0218698. pmid:32542001
- View Article
- PubMed/NCBI
- Google Scholar
51. Fraga RF, Rocha SM, Lima Neto IE. Impact of flow conditions on coliform dynamics in an urban lake in the Brazilian semiarid. Urban Water Journal. 2020;17(1):43–53.
- View Article
- Google Scholar
52. Shehane S, Harwood V, Whitlock J, Rose J. The influence of rainfall on the incidence of microbial faecal indicators and the dominant sources of faecal pollution in a Florida river. Journal of Applied Microbiology. 2005;98(5):1127–1136. pmid:15836482
- View Article
- PubMed/NCBI
- Google Scholar
53. Ouma S, Ngeranwa J, Juma K, Mburu D. Seasonal variation of the physicochemical and bacteriological quality of water from five rural catchment areas of lake victoria basin in Kenya. Journal of Environmental Analytical Chemistry. 2016;3(170):2.
- View Article
- Google Scholar
54. Shah VG, Dunstan RH, Geary PM, Coombes P, Roberts TK, Rothkirch T. Comparisons of water quality parameters from diverse catchments during dry periods and following rain events. Water Research. 2007;41(16):3655–3666. pmid:17428519
- View Article
- PubMed/NCBI
- Google Scholar
55. Lee TK, Oh EG, Yu HD, Ha KS, Yu HS, Byun HS, et al. Impact of rainfall events on the bacteriological water quality of the shellfish growing area in Korea. Korean Journal of Fisheries and Aquatic Sciences. 2010;43(5):406–414.
- View Article
- Google Scholar
56. Sibanda T, Chigor VN, Okoh AI. Seasonal and spatio-temporal distribution of faecal-indicator bacteria in Tyume River in the Eastern Cape Province, South Africa. Environmental Monitoring and Assessment. 2013;185(8):6579–6590. pmid:23242506
- View Article
- PubMed/NCBI
- Google Scholar
57. Topalcengiz Z, Strawn LK, Danyluk MD. Microbial quality of agricultural water in Central Florida. PloS One. 2017;12(4):e0174889. pmid:28399144
- View Article
- PubMed/NCBI
- Google Scholar
58. Vezzulli L, Oliveri C, Borello A, Gregory L, Kimirei I, Brunetta M, et al. Aquatic reservoir of Vibrio cholerae in an African Great Lake assessed by large scale plankton sampling and ultrasensitive molecular methods. ISME Communications. 2021;1(1):1–4.
- View Article
- Google Scholar
59. Mtui G, Nakamura Y. Physicochemical and microbiological water quality of Lake Sagara in Malagarasi wetlands. Journal of Engineering and Applied Science. 2006;1(2):174–180.
- View Article
- Google Scholar
60. St Laurent J, Mazumder A. The influence of land-use composition on fecal contamination of riverine source water in southern British Columbia. Water Resources Research Institute of the University of North Carolina Report. 2012;48(12).
- View Article
- Google Scholar
61. TBS. Tanzania Bureau of Standards; TZS 789:2003—drinking (potable) water—specification. In National Environmental Standards Compendium. 2003.
62. WHO. Guidelines for safe recreational water environments: Coastal and fresh waters: World Health Organization; 2003.
63. Crane PE, Silliman SE. Sampling strategies for estimation of parameters related to ground water quality. Groundwater. 2009;47(5):699–708. pmid:19459985
- View Article
- PubMed/NCBI
- Google Scholar
64. Bishop IJ, Warner S, van Noordwijk TC, Nyoni FC, Loiselle SJS. Citizen Science Monitoring for Sustainable Development Goal Indicator 6.3. 2 in England and Zambia. Sustainability. 2020;12(24):10271.
- View Article
- Google Scholar
65. Forrester G, Baily P, Conetta D, Forrester L, Kintzing E, Jarecki L. Comparing monitoring data collected by volunteers and professionals shows that citizen scientists can detect long-term change on coral reefs. Journal for Nature Conservation. 2015;24:1–9.
- View Article
- Google Scholar
66. Castilla EP, Cunha DGF, Lee FWF, Loiselle S, Ho KC, Hall C. Quantification of phytoplankton bloom dynamics by citizen scientists in urban and peri-urban environments. Environmental Monitoring and Assessment. 2015;187(11):690. pmid:26471276
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Niyoyitungiye L, Giri A, Ndayisenga M. Assessment of Coliforms Bacteria Contamination in Lake Tanganyika as Bioindicators of Recreational and Drinking Water Quality. South Asian Journal of Research in Microbiology. 2020:9–16.
View Article
Google Scholar

[2] View Article

[3] Google Scholar

[ref2] 2. Chen Kimirei IA. Demonstration Research on Comprehensive Water Quality Monitoring in the Lake Tanganyika Basin. CSTEC and NIGLAS, Nanjing, China. 2015.

[ref3] 3. Shen Q, Zhang L, Kimirei IA, Wang Z, Gao Q, Chen S, et al. Vertical physicochemical parameter distributions and health risk assessment for trace metals in water columns in eastern Lake Tanganyika, Tanzania. Journal of Oceanology and Limnology. 2019;37(1):134–145.
View Article
Google Scholar

[6] View Article

[7] Google Scholar

[ref4] 4. Stutter M, Langan S, Cooper R. Spatial and temporal dynamics of stream water particulate and dissolved N, P and C forms along a catchment transect, NE Scotland. Journal of Hydrology. 2008;350(3–4):187–202.
View Article
Google Scholar

[9] View Article

[10] Google Scholar

[ref5] 5. Zhang Y, Ma R, Hu M, Luo J, Li J, Liang Q. Combining citizen science and land use data to identify drivers of eutrophication in the Huangpu River system. Science of the Total Environment. 2017;584:651–664. pmid:28132775
View Article
PubMed/NCBI
Google Scholar

[12] View Article

[13] PubMed/NCBI

[14] Google Scholar

[ref6] 6. Cho S, Hiott LM, Barrett JB, McMillan EA, House SL, Humayoun SB, et al. Prevalence and characterization of Escherichia coli isolated from the Upper Oconee Watershed in Northeast Georgia. PloS One. 2018;13(5):e0197005. pmid:29738574
View Article
PubMed/NCBI
Google Scholar

[16] View Article

[17] PubMed/NCBI

[18] Google Scholar

[ref7] 7. Tornevi A, Bergstedt O, Forsberg B. Precipitation effects on microbial pollution in a river: lag structures and seasonal effect modification. PloS One. 2014;9(5):e98546. pmid:24874010
View Article
PubMed/NCBI
Google Scholar

[20] View Article

[21] PubMed/NCBI

[22] Google Scholar

[ref8] 8. Miguel-Chinchilla L, Heasley E, Loiselle S, Thornhill I. Local and landscape influences on turbidity in urban streams: a global approach using citizen scientists. Freshwater Science. 2019;38(2):303–320.
View Article
Google Scholar

[24] View Article

[25] Google Scholar

[ref9] 9. Gruber JS, Ercumen A, Colford JM Jr. Coliform bacteria as indicators of diarrheal risk in household drinking water: systematic review and meta-analysis. PloS One. 2014;9(9):e107429. pmid:25250662
View Article
PubMed/NCBI
Google Scholar

[27] View Article

[28] PubMed/NCBI

[29] Google Scholar

[ref10] 10. Ndlovu T, Le Roux M, Khan W, Khan S. Co-detection of virulent Escherichia coli genes in surface water sources. PloS One. 2015;10(2):e0116808. pmid:25659126
View Article
PubMed/NCBI
Google Scholar

[31] View Article

[32] PubMed/NCBI

[33] Google Scholar

[ref11] 11. Perkins TL, Clements K, Baas JH, Jago CF, Jones DL, Malham SK, et al. Sediment composition influences spatial variation in the abundance of human pathogen indicator bacteria within an estuarine environment. PloS One. 2014;9(11):e112951. pmid:25397595
View Article
PubMed/NCBI
Google Scholar

[35] View Article

[36] PubMed/NCBI

[37] Google Scholar

[ref12] 12. Gerba CP. Indicator microorganisms. Environmental Microbiology: Elsevier; 2009. p. 485–499.

[ref13] 13. Nkotagu HH. Lake Tanganyika ecosystem management strategies. Aquatic Ecosystem Health & Management. 2008;11(1):36–41.
View Article
Google Scholar

[40] View Article

[41] Google Scholar

[ref14] 14. Azanga OE. Land-use and land cover, sediment and nutrient hotspot areas changes in Lake Tanganyika Basin. African Journal of Rural Development. 2016;1(1):75.
View Article
Google Scholar

[43] View Article

[44] Google Scholar

[ref15] 15. Bergamino N, Horion S, Stenuite S, Cornet Y, Loiselle S, Plisnier P-D, et al. Spatio-temporal dynamics of phytoplankton and primary production in Lake Tanganyika using a MODIS based bio-optical time series. Remote Sensing of Environment. 2010;114(4):772–780.
View Article
Google Scholar

[46] View Article

[47] Google Scholar

[ref16] 16. Cohen AS, Palacios-Fest MR, Msaky ES, Alin SR, McKee B, O’Reilly CM, et al. Paleolimnological investigations of anthropogenic environmental change in Lake Tanganyika: IX. Summary of paleorecords of environmental change and catchment deforestation at Lake Tanganyika and impacts on the Lake Tanganyika ecosystem. Journal of Paleolimnology. 2005;34(1):125–145.
View Article
Google Scholar

[49] View Article

[50] Google Scholar

[ref17] 17. Donohue I, Irvine K. Seasonal patterns of sediment loading and benthic invertebrate community dynamics in Lake Tanganyika, Africa. Freshwater Biology. 2004;49(3):320–331.
View Article
Google Scholar

[52] View Article

[53] Google Scholar

[ref18] 18. Kalacska M, Arroyo-Mora JP, Lucanus O, Kishe-Machumu MA. Land cover, land use, and climate change impacts on Endemic Cichlid Habitats in Northern Tanzania. Remote Sensing. 2017;9(6):623.
View Article
Google Scholar

[55] View Article

[56] Google Scholar

[ref19] 19. Karamage F, Shao H, Chen X, Ndayisaba F, Nahayo L, Kayiranga A, et al. Deforestation effects on soil erosion in the Lake Kivu Basin, DR Congo-Rwanda. Forests. 2016;7(11):281.
View Article
Google Scholar

[58] View Article

[59] Google Scholar

[ref20] 20. Patterson G, Makin J. The state of biodiversity in Lake Tanganyika; a literature review. Chatham, UK: Natural Resource Institute; 1998. 1998.

[ref21] 21. Mziray P, Kimirei IA, Staehr PA, Lugomela CV, Perry WL, Trolle D, et al. Seasonal patterns of thermal stratification and primary production in the northern parts of Lake Tanganyika. Journal of Great Lakes Research. 2018;44(6):1209–1220.
View Article
Google Scholar

[62] View Article

[63] Google Scholar

[ref22] 22. Kelly B, Mtiti E, McIntyre PB, Vadeboncoeur Y. Stable isotopes reveal nitrogen loading to Lake Tanganyika from remote shoreline villages. Environmental Management. 2017;59(2):264–273. pmid:27826695
View Article
PubMed/NCBI
Google Scholar

[65] View Article

[66] PubMed/NCBI

[67] Google Scholar

[ref23] 23. Dickinson JL, Shirk J, Bonter D, Bonney R, Crain RL, Martin J, et al. The current state of citizen science as a tool for ecological research and public engagement. Frontiers in Ecology and the Environment. 2012;10(6):291–297.
View Article
Google Scholar

[69] View Article

[70] Google Scholar

[ref24] 24. Hughes RN, Hughes DJ, Smith IP. Citizen scientists and marine research: volunteer participants, their contributions, and projection for the future. Oceanography and Marine Biology: An Annual Review, Volume 52. 2014;52:257–314.
View Article
Google Scholar

[72] View Article

[73] Google Scholar

[ref25] 25. Hyder K, Wright S, Kirby M, Brant J. The role of citizen science in monitoring small-scale pollution events. Marine Pollution Bulletin. 2017;120(1–2):51–57. pmid:28477985
View Article
PubMed/NCBI
Google Scholar

[75] View Article

[76] PubMed/NCBI

[77] Google Scholar

[ref26] 26. Loiselle A S, Gasparini Fernandes Cunha D, Shupe S, Valiente E, Rocha L, et al. Micro and macroscale drivers of nutrient concentrations in urban streams in South, Central and North America. PloS One. 2016;11(9):e0162684. pmid:27662192
View Article
PubMed/NCBI
Google Scholar

[79] View Article

[80] PubMed/NCBI

[81] Google Scholar

[ref27] 27. Thornhill I, Loiselle S, Clymans W, van Noordwijk C. How citizen scientists can enrich freshwater science as contributors, collaborators, and co-creators. Freshwater Science. 2019;38(2):231–235.
View Article
Google Scholar

[83] View Article

[84] Google Scholar

[ref28] 28. Ceccaroni L, Bowser A, Brenton P. Civic education and citizen science: Definitions, categories, knowledge representation. Analyzing the role of citizen science in modern research: IGI Global; 2017. p. 1–23.

[ref29] 29. Gillett DJ, Pondella DJ, Freiwald J, Schiff KC, Caselle JE, Shuman C, et al. Comparing volunteer and professionally collected monitoring data from the rocky subtidal reefs of Southern California, USA. Environmental Monitoring and Assessment. 2012;184(5):3239–3257. pmid:21739280
View Article
PubMed/NCBI
Google Scholar

[87] View Article

[88] PubMed/NCBI

[89] Google Scholar

[ref30] 30. McGoff E, Dunn F, Cachazo LM, Williams P, Biggs J, Nicolet P, et al. Finding clean water habitats in urban landscapes: professional researcher vs citizen science approaches. Science of the Total Environment. 2017;581:105–116. pmid:28069307
View Article
PubMed/NCBI
Google Scholar

[91] View Article

[92] PubMed/NCBI

[93] Google Scholar

[ref31] 31. McKinley DC, Miller-Rushing AJ, Ballard HL, Bonney R, Brown H, Cook-Patton SC, et al. Citizen science can improve conservation science, natural resource management, and environmental protection. Biological Conservation. 2017;208:15–28.
View Article
Google Scholar

[95] View Article

[96] Google Scholar

[ref32] 32. Quinlivan L, Chapman DV, Sullivan T. Validating citizen science monitoring of ambient water quality for the United Nations sustainable development goals. Science of the Total Environment. 2020;699:134255. pmid:31683215
View Article
PubMed/NCBI
Google Scholar

[98] View Article

[99] PubMed/NCBI

[100] Google Scholar

[ref33] 33. Buytaert W, Zulkafli Z, Grainger S, Acosta L, Alemie TC, Bastiaensen J, et al. Citizen science in hydrology and water resources: opportunities for knowledge generation, ecosystem service management, and sustainable development. Frontiers in Earth Science. 2014;2:26.
View Article
Google Scholar

[102] View Article

[103] Google Scholar

[ref34] 34. Fuhrimann S, Stalder M, Winkler MS, Niwagaba CB, Babu M, Masaba G, et al. Microbial and chemical contamination of water, sediment and soil in the Nakivubo wetland area in Kampala, Uganda. Environmental Monitoring and Assessment. 2015;187(7):475. pmid:26122126
View Article
PubMed/NCBI
Google Scholar

[105] View Article

[106] PubMed/NCBI

[107] Google Scholar

[ref35] 35. APHA. Standard methods for the examination of water and wastewater, American Public Health Association, American Water Works association, Water Environment Federation. 1998.

[ref36] 36. White T. Monitoring a watershed: Nationwide turbidity testing in Australia. Volunt Monit. 1994;6:22–23.
View Article
Google Scholar

[110] View Article

[111] Google Scholar

[ref37] 37. Hunink J, Terink W, Contreras S, Droogers P. Scoping Assessment of erosion levels for the Mahale region, Lake Tanganyika, Tanzania. Unpubl FutureW Rep. 2015;148:1–47.
View Article
Google Scholar

[113] View Article

[114] Google Scholar

[ref38] 38. NBS. Tanzania National Bureau of statistics; 2012 population and housing census. 2013.

[ref39] 39. URT. Kigoma Socio-economic Profile, Ministry of Planning, Economy and Empowerment, Dar es Salaam. 2016.

[ref40] 40. Samikwa E, Kapute F, Tembo M, Phiri T, Holm RH. Identification of critical control points using water quality as an indicator of hygiene for artisanal fisheries on Lake Malawi. Lakes and Reservors: Research and Management. 2019;24(1):3–12.
View Article
Google Scholar

[118] View Article

[119] Google Scholar

[ref41] 41. Odongkara K, and Kyangwa I. Sanitation, fish handling and artisanal fish processing within fishing communities: socio-cultural influences.Jinja, Uganda,National Fisheries Resources Research Institute (NaFIRRI). 2005:46pp.

[ref42] 42. Byamukama D, Kansiime F, Mach RL, Farnleitner AH. Determination of Escherichia coli contamination with chromocult coliform agar showed a high level of discrimination efficiency for differing fecal pollution levels in tropical waters of Kampala, Uganda. Applied Environmental Microbiology. 2000;66(2):864. pmid:10653767
View Article
PubMed/NCBI
Google Scholar

[122] View Article

[123] PubMed/NCBI

[124] Google Scholar

[ref43] 43. Langenberg VT, Nyamushahu S, Roijackers R, Koelmans AA. External nutrient sources for Lake Tanganyika. Journal of Great Lakes Research. 2003;29:169–180.
View Article
Google Scholar

[126] View Article

[127] Google Scholar

[ref44] 44. Nkotagu HH, Athuman C. The limnology of the Lake Tanganyika sub catchment. Tanzania Journal of Science. 2007;33.
View Article
Google Scholar

[129] View Article

[130] Google Scholar

[ref45] 45. Christensen VG, Rasmussen PP, Ziegler AC, Jian X, editors. Continuous monitoring and regression analysis to estimate bacteria loads. Water Environment Federation, TMDL Science Issues Conference, St Louis, Missouri; 2001.

[ref46] 46. Loperfido J, Just CL, Papanicolaou AN, Schnoor JL. In situ sensing to understand diel turbidity cycles, suspended solids, and nutrient transport in Clear Creek, Iowa. Water Resources Research. 2010;46(6).
View Article
Google Scholar

[133] View Article

[134] Google Scholar

[ref47] 47. De Keukelaere L, Sterckx S, Adriaensen S, Bhatia N, Monbaliu J, Toorman E, et al. Coastal turbidity derived from PROBA-V global vegetation satellite. Remote Sensing. 2020;12(3):463.
View Article
Google Scholar

[136] View Article

[137] Google Scholar

[ref48] 48. Liu G, Wei J, Muthu B, Jackson Samuel RD. Chlorophyll-a concentration in the hailing bay using remote sensing assisted sparse statistical modelling. European Journal of Remote Sensing. 2020:1–12.
View Article
Google Scholar

[139] View Article

[140] Google Scholar

[ref49] 49. Tyler AN, Hunter PD, Spyrakos E, Groom S, Constantinescu AM, Kitchen J. Developments in Earth observation for the assessment and monitoring of inland, transitional, coastal and shelf-sea waters. Science of the Total Environment. 2016;572:1307–1321. pmid:26805447
View Article
PubMed/NCBI
Google Scholar

[142] View Article

[143] PubMed/NCBI

[144] Google Scholar

[ref50] 50. Bastaraud A, Perthame E, Rakotondramanga J-M, Mahazosaotra J, Ravaonindrina N, Jambou R. The impact of rainfall on drinking water quality in Antananarivo, Madagascar. PloS One. 2020;15(6):e0218698. pmid:32542001
View Article
PubMed/NCBI
Google Scholar

[146] View Article

[147] PubMed/NCBI

[148] Google Scholar

[ref51] 51. Fraga RF, Rocha SM, Lima Neto IE. Impact of flow conditions on coliform dynamics in an urban lake in the Brazilian semiarid. Urban Water Journal. 2020;17(1):43–53.
View Article
Google Scholar

[150] View Article

[151] Google Scholar

[ref52] 52. Shehane S, Harwood V, Whitlock J, Rose J. The influence of rainfall on the incidence of microbial faecal indicators and the dominant sources of faecal pollution in a Florida river. Journal of Applied Microbiology. 2005;98(5):1127–1136. pmid:15836482
View Article
PubMed/NCBI
Google Scholar

[153] View Article

[154] PubMed/NCBI

[155] Google Scholar

[ref53] 53. Ouma S, Ngeranwa J, Juma K, Mburu D. Seasonal variation of the physicochemical and bacteriological quality of water from five rural catchment areas of lake victoria basin in Kenya. Journal of Environmental Analytical Chemistry. 2016;3(170):2.
View Article
Google Scholar

[157] View Article

[158] Google Scholar

[ref54] 54. Shah VG, Dunstan RH, Geary PM, Coombes P, Roberts TK, Rothkirch T. Comparisons of water quality parameters from diverse catchments during dry periods and following rain events. Water Research. 2007;41(16):3655–3666. pmid:17428519
View Article
PubMed/NCBI
Google Scholar

[160] View Article

[161] PubMed/NCBI

[162] Google Scholar

[ref55] 55. Lee TK, Oh EG, Yu HD, Ha KS, Yu HS, Byun HS, et al. Impact of rainfall events on the bacteriological water quality of the shellfish growing area in Korea. Korean Journal of Fisheries and Aquatic Sciences. 2010;43(5):406–414.
View Article
Google Scholar

[164] View Article

[165] Google Scholar

[ref56] 56. Sibanda T, Chigor VN, Okoh AI. Seasonal and spatio-temporal distribution of faecal-indicator bacteria in Tyume River in the Eastern Cape Province, South Africa. Environmental Monitoring and Assessment. 2013;185(8):6579–6590. pmid:23242506
View Article
PubMed/NCBI
Google Scholar

[167] View Article

[168] PubMed/NCBI

[169] Google Scholar

[ref57] 57. Topalcengiz Z, Strawn LK, Danyluk MD. Microbial quality of agricultural water in Central Florida. PloS One. 2017;12(4):e0174889. pmid:28399144
View Article
PubMed/NCBI
Google Scholar

[171] View Article

[172] PubMed/NCBI

[173] Google Scholar

[ref58] 58. Vezzulli L, Oliveri C, Borello A, Gregory L, Kimirei I, Brunetta M, et al. Aquatic reservoir of Vibrio cholerae in an African Great Lake assessed by large scale plankton sampling and ultrasensitive molecular methods. ISME Communications. 2021;1(1):1–4.
View Article
Google Scholar

[175] View Article

[176] Google Scholar

[ref59] 59. Mtui G, Nakamura Y. Physicochemical and microbiological water quality of Lake Sagara in Malagarasi wetlands. Journal of Engineering and Applied Science. 2006;1(2):174–180.
View Article
Google Scholar

[178] View Article

[179] Google Scholar

[ref60] 60. St Laurent J, Mazumder A. The influence of land-use composition on fecal contamination of riverine source water in southern British Columbia. Water Resources Research Institute of the University of North Carolina Report. 2012;48(12).
View Article
Google Scholar

[181] View Article

[182] Google Scholar

[ref61] 61. TBS. Tanzania Bureau of Standards; TZS 789:2003—drinking (potable) water—specification. In National Environmental Standards Compendium. 2003.

[ref62] 62. WHO. Guidelines for safe recreational water environments: Coastal and fresh waters: World Health Organization; 2003.

[ref63] 63. Crane PE, Silliman SE. Sampling strategies for estimation of parameters related to ground water quality. Groundwater. 2009;47(5):699–708. pmid:19459985
View Article
PubMed/NCBI
Google Scholar

[186] View Article

[187] PubMed/NCBI

[188] Google Scholar

[ref64] 64. Bishop IJ, Warner S, van Noordwijk TC, Nyoni FC, Loiselle SJS. Citizen Science Monitoring for Sustainable Development Goal Indicator 6.3. 2 in England and Zambia. Sustainability. 2020;12(24):10271.
View Article
Google Scholar

[190] View Article

[191] Google Scholar

[ref65] 65. Forrester G, Baily P, Conetta D, Forrester L, Kintzing E, Jarecki L. Comparing monitoring data collected by volunteers and professionals shows that citizen scientists can detect long-term change on coral reefs. Journal for Nature Conservation. 2015;24:1–9.
View Article
Google Scholar

[193] View Article

[194] Google Scholar

[ref66] 66. Castilla EP, Cunha DGF, Lee FWF, Loiselle S, Ho KC, Hall C. Quantification of phytoplankton bloom dynamics by citizen scientists in urban and peri-urban environments. Environmental Monitoring and Assessment. 2015;187(11):690. pmid:26471276
View Article
PubMed/NCBI
Google Scholar

[196] View Article

[197] PubMed/NCBI

[198] Google Scholar

Figures

Abstract

Introduction

Material and methods

Study design

Study area

Recruitment and training of citizen science for water quality monitoring

Coliform monitoring

Turbidity monitoring

Rainfall

Population and activity drivers

Land use

Statistical analysis

Results

Temporal dynamics of coliform concentrations and turbidity

Potential drivers for coliform variabilities

Relationship between turbidity and coliforms

Comparison of citizen and professional scientist measurements

Discussion

Conclusions

Supporting information

S1 Table. Demographic characteristics of citizen scientists participated in coliform and turbidity monitoring in Lake Tanganyika.

S1 Graphical abstract.

Acknowledgments

References