Brian Irvine

Anticipating and managing future trade-offs and complementarities between ecosystem services

This paper shows how, with the aid of computer models developed in close collaboration with decision makers and other stakeholders, it is possible to quantify and map how policy decisions are likely to affect multiple ecosystem services in future. In this way, potential trade-offs and complementarities between different ecosystem services can be identified, so that policies can be designed to avoid the worst trade-offs, and where possible, enhance multiple services. The paper brings together evidence from across the Rural Economy and Land Use Programmes Sustainable Uplands project for the first time, with previously unpublished model outputs relating to runoff, agricultural suitability, biomass, heather cover, age, and utility for Red Grouse (Lagopus scotica), grass cover, and accompanying scenario narratives and video. Two contrasting scenarios, based on policies to extensify or intensify land management up to 2030, were developed through a combination of interviews and discussions during site visits with stakeholders, literature review, conceptual modeling, and process-based computer models, using the Dark Peak of the Peak District National Park in the UK as a case study. Where extensification leads to a significant reduction in managed burning and grazing or land abandonment, changes in vegetation type and structure could compromise a range of species that are important for conservation, while compromising provisioning services, amenity value, and increasing wildfire risk. However, where extensification leads to the restoration of peatlands damaged by former intensive management, there would be an increase in carbon sequestration and storage, with a number of cobenefits, which could counter the loss of habitats and species elsewhere in the landscape. In the second scenario, land use and management was significantly intensified to boost UK self-sufficiency in food. This would benefit certain provisioning services but would have negative consequences for carbon storage and water quality and would lead to a reduction in the abundance of certain species of conservation concern. The paper emphasizes the need for spatially explicit models that can track how ecosystem services might change over time, in response to policy or environmental drivers, and in response to the changing demands and preferences of society, which are far harder to anticipate. By developing such models in close collaboration with decision makers and other stakeholders, it is possible to depict scenarios of real concern to those who need to use the research findings. By engaging these collaborators with the research findings through film, it was possible to discuss adaptive options to minimize trade-offs and enhance the provision of multiple ecosystem services under the very different future conditions depicted by each scenario. By preparing for as wide a range of futures as possible in this way, it may be possible for decision makers to act rapidly and effectively to protect and enhance the provision of ecosystem services in the face of unpredictable future change.

Impact of prescribed burning on blanket peat hydrology

Fire is known to impact soil properties and hydrological flow paths. However, the impact of prescribed vegetation burning on blanket peatland hydrology is poorly understood. We studied 10 blanket peat headwater catchments. Five were subject to prescribed burning, while five were unburnt controls. Within the burnt catchments, we studied plots where the last burn occurred ∼2 (B2), 4 (B4), 7 (B7), or greater than 10 years (B10+) prior to the start of measurements. These were compared with plots at similar topographic wetness index locations in the control catchments. Plots subject to prescribed vegetation burning had significantly deeper water tables (difference in means = 5.3 cm) and greater water table variability than unburnt plots. Water table depths were significantly different between burn age classes (B2 > B4 > B7 > B10+) while B10+ water tables were not significantly different to the unburnt controls. Overland flow was less common on burnt peat than on unburnt peat, recorded in 9% and 17% of all runoff trap visits, respectively. Storm lag times and hydrograph recession limb periods were significantly greater (by ∼1 and 13 h on average, respectively) in the burnt catchments overall, but for the largest 20% of storms sampled, there was no significant difference in storm lag times between burnt and unburnt catchments. For the largest 20% of storms, the hydrograph intensity of burnt catchments was significantly greater than those of unburnt catchments (means of 4.2 × 10−5 and 3.4 × 10−5 s−1, respectively), thereby indicating a nonlinear streamflow response to prescribed burning. Together, these results from plots to whole river catchments indicate that prescribed vegetation burning has important effects on blanket peatland hydrology at a range of spatial scales.

Prediction of blanket peat erosion across Great Britain under environmental change

A recently developed fluvial erosion model for blanket peatlands, PESERA-PEAT, was applied at ten sites across Great Britain to predict the response of blanket peat erosion to environmental change. Climate change to 2099 was derived from seven UKCP09 future projections and the UK Met Office’s historical dataset. Land management scenarios were established based on outputs from earlier published investigations. Modelling results suggested that as climate changes, the response of blanket peat erosion will be spatially very variable across Great Britain. Both relative changes and absolute values of sediment yield were predicted to be higher in southern and eastern areas than in western and northern parts of Great Britain, peaking in the North York Moors of eastern England. Areas with high precipitation and low temperature were predicted to have low relative erosion changes and absolute sediment yield. The model suggested that summer desiccation may become more important for blanket peat erosion under future climate change, and that temperature was more dominant than precipitation in controlling long-term blanket peat erosion change. However, in the North York Moors, precipitation appeared to be more dominant in driving long-term erosion change. Land management measures were shown to provide a means to mitigate against the impacts of climate change on blanket peat erosion.

Erosion of Northern Hemisphere blanket peatlands under 21st-century climate change

Peatlands are important terrestrial carbon stores particularly in the Northern Hemisphere. Many peatlands, such as those in the British Isles, Sweden and Canada, have undergone increased erosion, resulting in degraded water quality and depleted soil carbon stocks. It is unclear how climate change may impact future peat erosion. Here we use a physically-based erosion model (PESERA-PEAT), driven by seven different global climate models (GCMs), to predict fluvial blanket peat erosion in the Northern Hemisphere under 21st-century climate change. After an initial decline, total hemispheric blanket peat erosion rates are found to increase during 2070-2099 (2080s) compared with the baseline period (1961-1990) for most of the GCMs. Regional erosion variability is high with changes to baseline ranging between -1.27 and +21.63 t ha-1 yr-1 in the 2080s. These responses are driven by effects of temperature (generally more dominant) and precipitation change on weathering processes. Low latitude and warm blanket peatlands are at most risk to fluvial erosion under 21st-century climate change.

A response to ‘Changes in water colour between 1986 and 2006 in the headwaters of the River Nidd, Yorkshire, UK: a critique of methodological approaches and measurement of burning management’ by Yallop et al

Yallop et al. (2011) question the methodological approach we used to determine the proportion of managed burning occurring in fifteen sub-catchments of the River Nidd, north-east England and our conclusion that managed moorland burning has no effect either on water colour in streams draining these sub-catchments or on the increase in water colour that has occurred over a 20 year period. In response, we defend the approach we used to determine the proportion of heather burning in these sub-catchments and the conclusions we drew in Chapman et al. (2010). Firstly, we would like to clarify that the aim of our paper, as stated in the introduction, was not to investigate the impact of moorland burning on catchment scale drainage water colour, but to investigate changes in stream water colour over a 20 year period and whether the timing and/or magnitude of colour release were the same for all sub-catchments. The main findings of Chapman et al. (2010) were: (i) Water colour increased in all sub-catchments, but percent increase varied considerably between sub-catchments ranging from 22 to 155%. Statistical analysis revealed that the subcatchments could be split into two ‘Types’ based on water chemistry and therefore dominant source of runoff. The chemistry of Type I streams was indicative of flow dominated by surface runoff or throughflow from peat (low concentrations of calcium, magnesium and silicon, typically mean pH \5), whereas the chemistry of Type II streams suggested that flow originated from deeper, or less peaty soils (typically mean pH [5.5, relatively high concentrations of calcium, magnesium and silicon). (ii) The largest proportional increases in water colour were observed in the sub-catchments that had the lowest mean annual water colour values in 1986, and in general, these were also the Type II catchments. In the discussion we considered a range of factors that may have accounted for the variability in water colour increase amongst the sub-catchments, one of which was managed heather burning. However, in the conclusion Chapman et al. (2010) speculated that the higher rate of increase in water colour in the Type II sub-catchments compared to the Type I sub-catchments is most likely related to changes in the ability of mineral horizons on the lower catchment slopes to adsorb dissolved organic carbon (DOC), and that these changes could be due to decreased acid sulphur deposition. As noted by Yallop et al. (2011), we did not provide a detailed description of how we assigned the proportion of managed burning for each sub-catchment within the paper and the possible sources of errors associated with our approach, and so take the opportunity to do so now. Managed heather burning was considered as an aggregated factor of observed burning activity derived from air photographs taken P. J. Chapman (&) S. M. Palmer B. J. Irvine G. Mitchell A. T. McDonald water@leeds, School of Geography, University of Leeds, Leeds LS2 9JT, UK e-mail: p.j.chapman@leeds.ac.uk

The PESERA-DESMICE Modeling Framework for Spatial Assessment of the Physical Impact and Economic Viability of Land Degradation Mitigation Technologies

This paper presents the PESERA-DESMICE integrated model developed in the EU FP6 DESIRE project. PESERA-DESMICE combines a process-based erosion prediction model extended with process descriptions to evaluate the effects of measures to mitigate land degradation, and a spatially-explicit economic evaluation model to evaluate the financial viability of these measures. The model operates on a grid-basis and is capable of addressing degradation problems due to wind and water erosion, grazing and fire. It can evaluate the effects of improved management strategies such as maintaining soil cover, retention of crop residues, irrigation, water harvesting, terracing and strip cropping. These management strategies introduce controls to various parameters slowing down degradation processes. The paper first describes how the physical impact of the various management strategies is assessed. It then continues to evaluate the applicability limitations of the various mitigation options, and to inventory the spatial variation in the investment and maintenance costs involved for each of a series of technologies that are deemed relevant in a given study area. The physical effects of the implementation of the management strategies relative to the situation without mitigation are subsequently valuated in monetary terms. The model pays particular attention to the spatial variation in the costs and benefits involved as a function of environmental conditions and distance to markets. All costs and benefits are added to a cash flow and a discount rate is applied. This allows a cost-benefit analysis to be performed over a comparative planning period based on the economic lifetime of the technologies being evaluated. It is assumed that land users will only potentially implement technologies if they are financially viable. After this framework has been set-up, various analyses can be made, including the effect of policy options on the potential uptake of mitigation measures and an analysis of where cost-effectiveness is highest. Apart from model description, we present case studies of the use of the framework to illustrate its functioning and relevance for policy-making.