Biodiversity-ecosystem service relationships in degraded and recovering ecosystems.

Abstract

Biodiversity loss is occurring at an unprecedented rate and most of this loss is due to human induced pressure. This loss in biodiversity had led to concerns that the provision of ecosystem services that humans depend upon might be negatively affected. As such much modern conservation science focusses on preserving biodiversity whilst protecting priority ecosystem services. However, there may be spatial and temporal trade-offs between these services and the biodiversity that is considered important. Characterisation of such the relationships between biodiversity and ecosystem services is vital in order to improve management and policies which aim to protect and restore both biodiversity and ecosystem services. The broad aims of the thesis were to explore biodiversity-ecosystem service relationships in (1) ecosystems invaded by non-native plant species and (2) tropical forests affected by human exploitation and disturbance. Specifically this thesis aimed to answer the questions: 1. What effect do non-native plant invasions have on aboveground carbon storage, belowground carbon storage, carbon sequestration, water quality and water provision? 2. How do changes in species richness affect this ecosystem service provision? 3. How do these changes relate to the woodiness and traits of invasive and native dominant species, and the type of ecosystem invaded? 4. What factors drive differences in residual stand damage, biomass loss and species richness change following selective logging? 5. After deforestation how long do carbon stocks and plant biodiversity take to recover in tropical forests? 6. Do carbon and plant biodiversity differ in their recovery rates? 7. Which areas are priorities for restoration of tropical carbon? All chapters in this thesis make use of large datasets that I collated from the literature and other authors in order to draw broad conclusions about trade-offs and relationships between services and biodiversity In the section concentrating on invasive species my results suggest that non-native invasive plants generally increase the storage of carbon, whilst reducing water quality and availability. This may indicate a fundamental trade-off between services where increased biomass of plants results in higher evapotranspiration and thus water loss, while also enhancing the carbon cycle and nitrogen production of microorganisms. In addition my results suggest that aboveground carbon storage increases as species richness is reduced, showing the opposite relationship to that shown in many biodiversity ecosystem functioning experiments. This is the first time any such relationship has been found between community change and ecosystem level impacts in the context of species invasions. However, it seems likely that this relationship depends on the identity and traits of the species, with invasions in open habitats by woody species likely to drive a negative relationship between richness change and biomass change with the opposite true when grassy species invade woodlands. This result presents a trade-off between conservation priorities that managers will need to consider. In Chapter 3 I investigated the possibility of predicting the impact of non-native invasive plant impacts on ecosystem services by using characteristics and functional traits of both invasive and native species. This work suggested that aboveground carbon storage is most easily predicted by traits and characteristics of native and non-native species, with few other ecosystem services well explained by models. Results suggested that transition from woody to non-woody dominant species resulted in most dramatic changes in aboveground carbon storage. However, interestingly aboveground carbon storage also tended to increase where native species were replaced by species of similar woodiness. Similarly, given that woodiness and size of species are related, there was a positive relationship between the invasive species height and increases in aboveground carbon storage. However, all other ecosystem services were poorly predicted by species traits and characteristics. This work suggests that the most dramatic changes in carbon storage may result from shifts in ecosystems that resemble regime shifts. Future work addressing invasive species from this perspective is warranted as many invasions resemble such shifts. In Chapter 4 I investigated the relationships between logging intensity and methods and residual stem damage, biomass loss and species richness change in tropical logged forests. Many syntheses of the logging literature have made little distinction between logged sites, and only one has explored any of the mechanisms that may drive heterogeneity in logging impacts. This is particularly surprising given that Reduced Impact Logging (RIL) has been implemented relatively widely principally to reduce carbon loss from logged forests. My results from this chapter suggest that the principal driver of logging impacts is the intensity at which logging is carried out, showing broadly negative relationships with biomass and tree species richness change and a positive relationship with residual stem damage. Interestingly, RIL appeared to reduce residual stem damage slightly but evidence for this effect was weaker in other analyses. These analyses also suggest a slight increase in tree species richness at low logging intensities, showing some similarities to intermediate disturbance hypothesis type relationships. This is suggestive of a complex relationship between tree species richness and biomass changes during logging that deviated substantially from that suggested in grassland biodiversity-ecosystem function experiments. This is as far as I know the first time this relationship has been suggested in the context of logged forests. The result from this chapter also suggest that there is weak support that RIL reduces logging damage at low intensities but little evidence that this is reflected by changes in biomass. Further studies are needed to discern the effect of RIL over a wide range of logging intensities. Chapter 5 investigates tropical forest recovery following agricultural clearance. In this chapter I aimed to identify the recovery times of different above and belowground carbon pools and tree and epiphyte species richness as well as tree species composition using studies that had paired mature forest sites as comparators. Surprisingly this chapter represents the first attempt to generalise about this recovery rate. The results suggest that following clearance carbon and species richness of plants recovers relatively quickly (<100 years), but species indicative of old forests are rarely present in recovering forests and show few signs of recovery. Thus, while carbon recovery goals may be achievable full recovery of plant biodiversity may require centuries. This slow recovery may be aided by active restoration. Finally in Chapter 6 I investigated which areas should be considered as priorities when restoring tropical forests for carbon storage and bird biodiversity. In this chapter I found evidence of spatial trade-offs between carbon storage and bird species recovery. Empirical models suggested that carbon is accumulated most rapidly in forests with long growing seasons, while probability of bird species presence was primarily driven by habitat specificity, range size and forest cover. Model projections suggested that areas that should be considered a priority for restoration targeting carbon storage are found in the wet tropics while priorities for restoration of bird biodiversity are found in mountainous areas. These analyses indicated that there was no relationship between the two goals, but that by using model projections it was possible to identify areas that maximised both. In summary work in this thesis provides the best synthesis of the relationships between biodiversity and ecosystem services in the context of non-native invasive plants, and selective logging and recovery from tropical forest clearance to date. This is of particular value because such relationships have rarely been explored in these contexts despite widespread and of global importance for conservation.