Assessing the impact of socio-economic development and climate change on faecal indicator bacteria in the Betna River, Bangladesh

Consumption of water that is contaminated with pathogens still causes high numbers of death and disease. Understanding the factors that influence the dynamic distribution of waterborne pathogens is important, as this will help understanding improvements and possible solutions. Such understanding is particularly important in a developing country like Bangladesh, where large proportions of the population often have little or no access to clean water. Despite the high relevance for public health, few studies currently exists on the fate and transport of pathogens and the so-called Faecal Indicator Bacteria (FIB, e.g. <em>E. coli, </em>enterococci) in (sub)tropical systems. FIB are susceptible to shifts in water flow and quality. The predicted increases in rainfall and floods due to climate change will exacerbate the faecal contamination scenarios. This could be further compounded by the rapid change in socio-economic conditions (population growth, urbanization, sanitation and agricultural management) in the developing countries. Therefore, to reduce future health risks, understanding the influence of changes in socio-economic conditions and climate on microbial dynamics is important. Very few studies have quantified the relationship between the waterborne pathogens/FIB concentrations and climate and socio-economic changes. In this study a process-based model was developed and a scenario analysis was performed based on the new combined climate and socio-economic changes scenarios, to assess the present and future river hydrodynamics, FIB sources, die-off processes and concentrations. We used FIB, because measuring FIB are cheaper than pathogens. FIB are usually not pathogenic but their presence indicates the likely presence of waterborne pathogens. These pathogens are expected to respond to climate change in a comparable way to FIB. The present study is based on the Betna River basin in southwestern Bangladesh, where faecal contamination is not monitored and very little knowledge exists on the distribution of contaminants. First of all, FIB concentrations of the river water were measured to identify the river’s faecal contamination levels that can be used to validate the water-quality model. In the study area, wastewater is not treated and this untreated wastewater is discharged directly into the river. This is evident from the measured FIB data. In 88% of the <em>E. coli </em>and all enterococci samples, the USEPA bathing water quality standards were violated (Chapter 2). Such violation indicates potential health risks associated with the use of the river water for domestic, bathing and irrigation purposes. The correlation between environmental variables (water temperature, precipitation and salinity) and FIB concentrations was also determined. A positive correlation was found with water temperature and precipitation, and a negative correlation with salinity. The positive correlation with temperature is due to the co-occurrence of high summer temperature with abundant monsoon rainfall. The positive correlation with precipitation can be explained by the increased runoff from agricultural lands and urban areas. This runoff contains many bacteria. In the study area, during the rainy season (July to September) precipitation increases and as a result water salinity decreases. The observed negative correlation with salinity is more likely due to the typical weather patterns during the rainy season when low salinity coincides with increased precipitation and high temperature, than to salinity dependent die-off of bacteria. A regression model was applied that explained almost half of <em>E. coli </em>and enterococci variability in river water. This, however, only considers water temperature and precipitation (Chapter 2). Then, the present and future hydrodynamics of the river were simulated using a two dimensional hydrodynamic model (MIKE 21 FM). Although the main goal of this thesis is to assess the river’s present and future FIB concentrations, the reasons for this hydrodynamic modelling are twofold. Firstly, outputs of the hydrodynamic model are used as input into the water-quality model (Chapter 4). Secondly, hydrodynamics (i.e. water level and discharge) are simulated because increased water level and discharge together with sea level rise stimulate floods in the river basin. These floods are related to outbreaks of waterborne diseases. The modelled results corresponded very well with the measured water levels and discharges. The model was applied to simulate baseline and future water levels and discharge for Representative Concentration Pathway RCP4.5 and RCP8.5 scenarios using bias-corrected downscaled data from two climate models (IPSL-CM5A and MPI-ESM). The model results showed an expected increase in water level up to 16% by the 2040s and 23% by the 2090s (Chapter 3). The monsoon daily maximum discharge was expected to increase up to 13% by the 2040s and 21% by the 2090s. These model results also showed that the duration of the water level above the danger level and extreme discharge periods can increase by half a month by the 2040s and over a month by the 2090s. The coincidence of the water danger level with extreme discharge may cause disastrous floods in the study area. Next, the hydrodynamic model was coupled with a water-quality module (ECOLab). The fate and transport of FIB was simulated, the influence of different processes tested and the contribution from different sources to the total contamination quantified (Chapter 4). The model outputs corresponded very well with the measured FIB data. The present river microbial water quality based on measured and simulated results indicated, once again, noncompliance with bathing water standards. Primary and secondary levels of wastewater treatment were not sufficient to reach the standards most of the time, and discharges from sewer drains and incoming concentrations from the upstream boundary were found to be a major cause of water contamination. Tide, wind and diffuse sources (urban and agricultural runoff) contributed little. The high FIB inputs from the upstream open boundary come from untreated point source discharges from upstream urban areas and accumulation of diffuse contaminants from the large upstream areas. Therefore, this study underlines the need for establishment of wastewater treatment plants both in the studied basin and upstream urban areas. This study provides insight into bacterial fate and transport mechanisms, contribution of different sources to the faecal contamination and applicability of wastewater treatment in a river of a subtropical developing country where this type of study is lacking. Uncertainties are related to the lack of high temporal resolution measured FIB data and the lack of available data for contaminant loads from septic tank leakages, open defecation and sediment resuspension. However, the model well captured the measured FIB variability, suggesting that it can be applied for microbial water quality assessments in other watersheds of the world with similar characteristics. The developed model could be an ideal tool to forecast future impacts of climate and socioeconomic changes on FIB fate, transport and dynamics. Finally, future FIB concentrations were simulated using the coupled hydrodynamic and microbial model (MIKE 21 FM-ECOLab) and scenario analysis (Chapter 5). Scenarios have been developed building on the most recent Shared Socio-economic Pathways (SSPs) and Representative Concentration Pathways (RCPs) scenarios from the Intergovernmental Panel on Climate Change (IPCC). We developed a baseline scenario (October 2014–September 2015) reflecting the current conditions and two future scenarios, S1 (sustainability scenario) and S2 (uncontrolled scenario) mimicking different future developments of socio-economic (population, urbanization, sanitation, wastewater treatment development, land use) and climate-change factors (temperature, precipitation and sea-level rise). In S1 RCP4.5 was combined with socio-economic scenarios SSP1, and for S2 RCP8.5 was combined with SSP3 (S2). Assumptions on sanitation, waste water treatment and agricultural management in line with the storylines were made to quantify future changes in FIB concentrations and consequent health risk. Different future scenarios were found to have substantial impact on FIB concentrations in the river. By the 2090s, FIB concentrations are expected to decrease by 98% or increase by 75% for the sustainability scenario and uncontrolled scenario respectively. An uncontrolled future resulted in a deterioration of microbial water quality due to socio-economic developments, such as higher population growth, land-use change and increased sewage discharges and changes in rainfall patterns. Microbial water quality strongly improved under a sustainable climate and improved sewage treatment. FIB concentrations were much more sensitive to changes in socio-economic factors than to changes in climatic factors. This underlines the importance of socio-economic factors in assessing and improving microbial water quality. The results show the importance of improvements in sanitation and wastewater treatment in the Bangladeshi Betna River basin to ensure that future FIB concentrations in the river comply with the US-EPA bathing water quality standards. Major investments to construct wastewater treatment plants are necessary to compensate for the population growth and increased the volume of wastewater treatment. Although the current level of contamination is already too high, without wastewater treatment the water quality will further deteriorate. The thesis assesses the present and future FIB dynamics in the Betna River through sampling, statistical and process-based modelling, and scenario analysis. The results contribute to increase the knowledge base on the dynamic distributions of the FIB in surface water in a developing country and in a subtropical system, where this type of study is lacking. It also reduces the knowledge gaps regarding future flooding scenarios at the local scale. While some earlier studies focused on only assessing climate-change impacts on microbial water quality, this study for the first time assessed the influence of combined climate and socio-economic scenarios (using scenarios based on the new SSP-RCP scenario matrix) on river FIB concentrations. This combined modelling and scenario approach enables the assessment of faecal contamination sources and dynamics at present and in the future. The developed model and scenario analysis approach provides a basis for the water managers to reduce the widespread faecal contamination and the risks of waterborne disease outbreaks, which are still a leading cause of deaths in developing countries.