Weak evidence of improvement in ecological indicators of water quality following country-wide mitigation measures of intensively managed agriculture catchments

The improvement of watercourses' ecological quality is required by several legal landmarks and is an objective for sustainable agriculture. Here we assessed the potential of sets of mitigation measures proposed to improve water quality in Irish river catchments intensively used for agriculture. We used a spatially replicated and long-term dataset of macroinvertebrates and diatoms in six intensively managed agricultural catchments. Sampling was conducted in small-order streams in the spring and autumn from 2009 to 2021 as part of the Irish Agricultural Catchments Programme. Mitigation measures of diffuse pollution and erosion in different processes of the nutrient-transfer continuum were adopted during the monitoring period as Good Agriculture Practices (GAP). A strengthening of GAP aiming to reduce the amount of nutrients and sediments entering watercourses took in effect in 2018. We used this temporal landmark to assess the early efficacy of this set of measures splitting the time series of the monitoring before and after GAP measures strengthening. We explored the potential effects of GAP with interrupted-time series (ITS) analyses presenting time as a discrete variable describing the first to the last sampling, and nominal variables describing the years before and after GAP strengthening. The effect of an intervention can be observed in ITS by an immediate change in the average value of the response after the intervention and/or by a change in the temporal trend of the response variable variation. The influence of climate, spatial and temporal autocorrelation in the ITS models was assessed with the scores of a Principal Component Analysis describing climate as a co-variable, site identity as a random-effects variable, and a first-order temporal autocorrelation function in linear mixed-effects models. The response variables were the macroinvertebrates taxa richness, Ephemeroptera, Plecoptera and Trichoptera (EPT) richness, diatoms species richness, and the relative abundance of sensitive or tolerant diatoms in these ITS models. The GAP implementation often correlated with immediate increases in the average macroinvertebrates or EPT taxa richness, but the temporal trends of decreasing taxa richness became steeper after GAP (Figs 1, 2). Diatoms species richness increased through time independently of GAP in one catchment (Ballycanew), while in another catchment (Timoleague) GAP correlated with an increase in the average diatoms species richness (Fig. 3). The relative abundance of diatoms which grows better under a low quantity of nutrients (sensitive diatoms) tended to increase through time in three catchments, while the GAP implementation period correlated with an immediate decrease in the average relative abundance of these diatoms. However, the GAP implementation period correlated with a change of direction of the temporal trend in the Timoleague catchment, with the relative abundance of sensitive diatoms decreasing through time after GAP (Fig. 4). GAP correlated with an immediate increase in the relative abundance of diatoms which grow better under a high quantity of nutrients (tolerant diatoms) in Ballycanew catchment. For two catchments (Castledockerell and Timoleague), the temporal trend of variation in relative abundance for tolerant diatoms changed after GAP (Fig. 5). In sum, the mitigation measures taken country-wide present little evidence of improving the diversity and relative abundance of biological indicators after 2018. Our results indicate that improving Irish water quality needs a closer interaction with farmers engaging them to lessen the loss of nutrients, and sediments to watercourses and to restore connectivity and habitat in agricultural catchments. For this, educational activities such as periodic training, and events of knowledge transfer between science and practice are essential to increase adherence to mitigation measures already proposed and to weaken barriers to adopting those measures. Furthermore, additional measures aiming to improve connectivity between habitats within catchments such as providing continuous riparian vegetation, improving the vegetation structure for those already present or targeting priority areas to create these areas can be adopted to improve the dispersal and persistence of species throughout the catchments.