Thresholds of biodiversity and ecosystem function in a forest ecosystem undergoing dieback

www.nature.com/scientificreports OPEN Received: 4 January 2017 Accepted: 7 June 2017 Published online: 28 July 2017 Thresholds of biodiversity and ecosystem function in a forest ecosystem undergoing dieback P. M. Evans 1, A. C. Newton1, E. Cantarello1, P. Martin Barsoum4, J. E. Cottrell4, S. W. A’Hara4 & L. Fuller5 1 , N. Sanderson2, D. L. Jones3, N. Ecological thresholds, which represent points of rapid change in ecological properties, are of major scientific and societal concern. However, very little research has focused on empirically testing the occurrence of thresholds in temperate terrestrial ecosystems. To address this knowledge gap, we tested whether a number of biodiversity, ecosystem functions and ecosystem condition metrics exhibited thresholds in response to a gradient of forest dieback, measured as changes in basal area of living trees relative to areas that lacked recent dieback. The gradient of dieback was sampled using 12 replicate study areas in a temperate forest ecosystem. Our results provide novel evidence of several thresholds in biodiversity (namely species richness of ectomycorrhizal fungi, epiphytic lichen and ground flora); for ecological condition (e.g. sward height, palatable seedling abundance) and a single threshold for ecosystem function (i.e. soil respiration rate). Mechanisms for these thresholds are explored. As climate-induced forest dieback is increasing worldwide, both in scale and speed, these results imply that threshold responses may become increasingly widespread. The living world is currently experiencing an unprecedented period of environmental change1–4. In recent decades, human-derived actions such as carbon emission, introduction of species and large-scale land transformations (e.g. urban and agricultural expansion) have become pervasive throughout the biosphere. Impacts of human activity have become so widespread and intrusive that a new geological epoch, the Anthropocene, has been proposed5. Human actions have influenced the functioning of the Earth system to such an extent that the consequences could be detrimental or even catastrophic for human society1–4. This is reflected in development of the planetary boundaries concept, which suggests that if specific thresholds of environmental change are transgressed, there may be increased risks to human wellbeing or to resilience of the whole Earth system2, 3. The concept of planetary boundaries, together with allied concepts such as resilience2, 3, depends on the existence of ecological thresholds. Such thresholds are defined as points or zones where relatively rapid change occurs from one ecological condition to another6, and are characterised by a non-linear response of an ecosystem property to a controlling variable that increases linearly7. If thresholds occur in nature, a slight increase in disturbance intensity or frequency could cause a disproportionate change in an ecosystem property. Such changes could include the loss of biodiversity crucial for ecosystem function8 and the loss of regulatory ecosystem services on which humans depend9. Moreover, a threshold in one ecosystem property could sequentially disrupt the self-organising networks that govern local dynamics of other systems10, and could potentially cause unpredictable responses at the scale of whole Earth system dynamics3, 6, 11. There is a need to avoid crossing such thresholds to enable ecological systems, and their associated socio-economic systems, to be maintained in the future12. Ecological thresholds are thought to be attributable to shifts in the relative strength of balancing (i.e. negative) and reinforcing (i.e. positive) feedback loops that influence the dynamics of an ecosystem13. For example, in many terrestrial ecosystems, low water availability acts to regulate the growth of plants. Conversely, if water availability increases by a sufficient amount, the biomass and complexity of vegetation can increase, which can further increase water availability by modifying the water cycle14, 15. 1 Centre for Conservation Ecology and Environmental Sciences, Faculty of Science and Technology, Bournemouth University, Poole, BH12 5BB, UK. 2Botanical Survey and Assessment, 3 Green Close, Woodlands, Southampton, Hampshire, SO40 7HU, UK. 3School of Environment, Natural Resources and Geography, Bangor University, Gwynedd, LL57 2UW, UK. 4Forest Research, Alice Holt Lodge, Farnham, Surrey, GU10 4LH, UK. 5Biological and Environmental Sciences, University of Stirling, Stirling, FK9 4LA, UK. Correspondence and requests for materials should be addressed to P.M.E. (email: ) SCiEnTiFiC REPOrTS | 7: 6775 | DOI:10.1038/s41598-017-06082-6 1 www.nature.com/scientificreports/ Figure 1. Distribution of broadleaved woodland (green), occurrence of dieback (red) and location of each of the 12 study areas (pink dots) in the New Forest, in southern England. Map was made using ArcMap 10.1 (http://desktop.arcgis.com/en/arcmap/). Despite the perceived global importance of ecological thresholds, supporting evidence is largely theoretical7, 16, and the issue is the focus of major scientific debate17, 18. Supporting empirical evidence from field situations is severely limited6, 19, and is primarily available for aquatic systems20–22. Field evidence for ecological thresholds resulting from environmental change is particularly lacking in temperate woodland ecosystems that are not governed by fire6, 23. This research therefore aimed to test the hypothesis that threshold responses exist in measures of (1) biodiversity, (2) ecosystem function and (3) ecosystem condition within a terrestrial ecosystem, specifically temperate forest. To test this hypothesis, we examined a beech-dominated forest that is currently undergoing large-scale dieback in response to environmental change, as revealed through analysis of long-term monitoring data24. Methods Study area. We carried out this study in the New Forest National Park (NP), which covers an area of 57,100 ha situated in southern England (longitude: 1°17′59″ to 1°48′8″ W, Latitude: 50°42′19″ to 51°0′17″ N) (Fig. 1). The Forest consists of a mosaic of heathland, mire, grassland and coniferous and broadleaf woodland (8,472 ha) ecosystems. These woodlands are dominated by beech (Fagus sylvatica), often occurring with oak (Quercus robur) and birch (Betula pendula), and typically with holly (Ilex aquifolium) in the understorey25. The local climate is oceanic and temperate, with a mean annual maximum temperature of 14.8 °C and annual rainfall of 835.2 mm, based on data available between 1981 and 201026. The Park contains the largest area of semi-natural vegetation in lowland Britain27, 28, and is of exceptional importance for biodiversity conservation29. The New Forest is also characterised by high densities of large herbivores, including livestock and deer, reflecting its history as a Royal hunting reserve27. Experimental design. A geographic information system (GIS) (ArcGIS 10.1) was utilis (...truncated)