Climate change and eutrophication risk thresholds in English rivers

Climate change is expected to alter water quality in rivers, but where and when this may happen is uncertain. This report describes a study of projected response in the amount of algal plant growth (phytoplankton biomass). Increasing algal growth is one of the ecological manifestations of eutrophication in slow flowing rivers, where the water starts to resemble a green soup. Eutrophication is a process in which too much nutrient in water causes algae and higher plants to grow excessively. Eutrophication alters the quality of the water and how it can be used. Phytoplankton (suspended algae) is considered to be a useful indicator of eutrophication in standing freshwaters and can also be useful as one measure of impacts in rivers, particularly slow flowing rivers. Excess algal growth can result in blooms that eventually die off. The disruption of dissolved oxygen dynamics in the water column may, in turn, have adverse impacts on fish and macroinvertebrates. The onset and decline of algal blooms is measured by the concentration of chlorophyll (a green pigment in algae) in the water. In this context, algal bloom risk – and the risk of negative eutrophication impacts in the lower reaches of rivers – is identified through observations of threshold chlorophyll concentrations. Exceedence of a chlorophyll concentration threshold is not by itself used in the diagnosis of river eutrophication but can be used as a proxy for algal blooms for understanding and modelling risk. The future risk of eutrophication impact, including algal blooms, is affected by changes in the concentration of nutrients from altered river flow and changes in phosphorus inputs from a range of sources. An earlier study (Phase 1 of this project) demonstrated that climate change impacts on river flow would increase phosphorus concentrations by 2050 and beyond. However, climate-driven changes in river temperature regime and light, and plant responses to these, are also important in altering the future risk of excess algal growth. This report considers these aspects. The first step was to identify the variables that control eutrophication and the thresholds in these variables which determine the potential for algal blooms. Algal blooms tend to occur only in rivers with a residence time (the time water takes to travel from an upstream distance to a site) of over 4 days. Below 4 days, blooms are rare. Such long residence times in the UK tend to occur in canals, and slow flowing and shallow gradient rivers (often in their lower reaches). Using this residence time threshold of 4 days, a total of 26 sites in England on 24 different rivers with available data for analysis of trends were identified out of the 115 sites from Phase 1. Water quality data were used to identify the ranges of river flow and water temperature within which algal blooms were measured (as determined by chlorophyll concentration) for each site. Site-specific thresholds were identified from plots of variables of water quality against chlorophyll concentration. In this study, a chlorophyll threshold of 30µgl-1 indicated the onset of an algal bloom for most rivers. Thresholds ranged between 15µgl-1 and 100µgl-1 . For larger rivers, with higher chlorophyll levels (such as the Thames), the thresholds for algal blooms are higher. A phosphorus threshold of 30µgl-1 was selected for all sites, based on understanding developed through nutrient limitation experiments across a range of UK rivers in other studies. A sunlight duration threshold of 65W/m2 /day was chosen for all the sites based on a minimum of at least 3 hours of full sunshine per day over ~3 consecutive days (derived from earlier work). A bloom is likely to occur if all thresholds are met at the same time. These are called bloom risk days and they represent overall risk based on all measured variables. A spreadsheet model was developed and applied to the 26 sites. The model used daily estimates of controlling variables (phosphorus concentrations, river flow, water Climate change and eutrophication risk thresholds in English rivers v temperature and sunlight duration) from 1951 to 2098 to estimate when the derived thresholds for each variable were met and likely to cause an algal bloom. Phosphorus concentration estimates from earlier work were used under current wastewater treatment conditions and under an improved wastewater treatment scenario. Bloom risk days (when the river flow, water temperature, sunshine duration and phosphorus concentration thresholds for algal growth were all met) increased between the baseline period (1961 to 1990) and the 2050s future period (2040 to 2069). The median increase is about 8 days across all sites from about 50 in the baseline period, although the maximum increase is up to 15 days. The change in risk is variable by the 2080s (2070 to 2098), with about 50% of sites showing reduced risk relative to the baseline period, resulting in a median increase of about 4 days and a maximum of up to 16 days. Analysis of the number of threshold days for each individual driver indicates that phosphorus thresholds are met most days of the year and that phosphorus concentrations do not prevent bloom development except at one site. Phosphorus management strategies may therefore not be effective in reducing the risk of algal blooms occurring in slow flowing rivers, an observation confirmed by the fact only 3 sites showed a reduction in risk using an improved phosphorus treatment scenario. There is more variability in the number of days the other thresholds are met, resulting in a varying pattern of risk between sites and time periods. After phosphorus concentration thresholds, river flow thresholds are most frequently met. Sunlight duration and water temperature thresholds are least often met. The interaction between flow variability, water temperature and sunlight duration would appear to determine the variability that emerges by the 2080s. The role of water temperature and sunlight duration seems to be significant in both limiting the number of days all thresholds are met and in controlling the timing of attainment of all thresholds, with both thresholds tending to be exceeded later in the year than those for river flow and phosphorus concentration. With the lowest number of threshold days at the greatest number of sites, exposure to sunlight may be the most important factor in preventing algal blooms. There is considerable uncertainty in the estimation of future water temperature, which was derived from air temperature using simple regression methods. This may result in a variable estimate of bloom risk days that requires further exploration with more reliable projections of future water temperature. A better way of estimating water temperature would really help to model future water quality. These results suggest that management strategies focusing on reducing sunlight and thermal interactions (both through river shading by trees) may be particularly effective in reducing the risk of blooms on some rivers in the future. This could be explored using the spreadsheet model developed for this project. Whilst phytoplankton blooms tend to be observed in lowland reaches of English rivers, the approach applied here is independent of this, is equally applicable anywhere, and has potential for use in an approach for assessing eutrophication in slow flowing rivers. It would also be useful to identify more sites across England at which residence time thresholds are met in order to assess potential vulnerability to eutrophication.