Understanding drivers of long-term change in semi-natural habitats.

Abstract

Biodiversity has declined considerably across the globe and this is largely due to the destruction and degradation of habitats, which is predicted to continue in the future. Understanding the dynamics and mechanisms behind habitat loss and degradation can help predict future change and plan potential ameliorative actions, such as restoration. Long-term change studies have proved exceptionally valuable for understanding these drivers, though few studies examine more than two snapshots in time. As a result the current knowledge of these drivers of change, their trends and the resulting responses is deficient. This thesis examines the long-term drivers of habitat loss and quality in southern England over multiple time periods for a range of semi-natural habitats. The research questions are as follows; (i) What are the spatio-temporal patterns and drivers of habitat loss? (ii) Does habitat loss lead to extinction debts across a range of habitat types? (iii) What are the temporal patterns and drivers of vegetation change in calcareous grasslands? Chapters 2 and 3 address the first research question by exploring the dynamics of habitat change over time for a rural county in southern England. Historical surveys and contemporary spatial datasets were used to construct a habitat time-series dataset for over 3700 sites across the county between 1930 and 2015. This dataset was then used to model landscape maps for the whole county in 1950 and 1980 using the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) Scenario Generator tool. These chapters revealed both the spatial and temporal patterns of loss, which were largely non-linear for the majority of semi- natural habitats between 1930 and 2015. The greatest losses occurred between 1930 and 1980, which coincided with the period of most rapid agricultural intensification in the UK. However, even after this period semi-natural habitats were still being lost, despite so few sites remaining. Statutory protection was important for retaining semi-natural habitats, which provides evidence for their effectiveness and the need for continued protection in the future. The fourth chapter utilises the datasets produced in the previous two chapters to test for plant community extinction debts (delayed species extinctions following habitat degradation) in three of the semi-natural habitats (calcareous grassland, heathland and broadleaved woodland), thus addressing the second research question. A key knowledge gap is addressed by comparing three different extinction debt methods. These three species-area relationship methods differ in their data requirements, with the first two using information on past and current habitat area alongside current species richness, whilst the last method also requires data on past species richness. The most data-intensive method identified extinction debts across all three habitats for specialist species, whilst the other methods did not. This suggests that extinction debts may be under-reported, which has important implications for employing conservation action. Chapters 5 and 6 address the third research question by focusing on long-term change and drivers of habitat quality using calcareous grassland as a case study. Both chapters used vegetation survey data from National Nature Reserves (NNR) spanning between the 1970s and 2010s. Chapter 5 used vegetation and soil data collected at Parsonage Down at three time points (1970, 1990 and 2016) and found declines in species richness, diversity and shifts in species composition across the survey period. Nitrogen deposition may have contributed towards community changes between 1970 and 1990, as indicated by the change in soil properties and the associated decline in species diversity. However, between 1990 and 2016 vegetation change is likely to be largely attributable to a reduction in grazing pressure, indicated by the increase in taller species and a decrease in grazing tolerance. Chapter 6 examined vegetation change alongside detailed grazing records across multiple time periods at Martin Down. Although a significant shift in the grazing regime was detected, whereby grazing previously occurred across all seasons to now predominantly occurring in autumn and winter, the vegetation communities remained largely consistent. This suggests that these vegetation communities are robust to changes in seasonal grazing providing that sufficient grazing pressure is provided within the year. The findings presented in this thesis demonstrate the importance of long-term data and the value of analysing multiple intermediate time periods to identify trends in drivers of change and their responses.