Evaluating ecosystem service assessments in multifunctional landscapes in New Zealand

Abstract

Sustaining ecosystem function and human livelihoods is a management focus in multifunctional landscapes. Multifunctional landscapes provide a range of provisioning (e.g. food production), regulating (e.g. pollination), cultural (e.g. recreational opportunities), and supporting (e.g. habitat for biodiversity) ecosystem services (ES). Many methods for assessing ES supply have been proposed, making it challenging to identify which ES assessment approach is most appropriate given the data available, the research question, and policy objectives. Assessing the supply, distribution and spatial interactions between multiple ES across different land-use scenarios is important if we are to understand how different management objectives or land-use changes will affect the supply of key ES. In many assessment approaches ES are represented as static, but many services are spatially and temporally dynamic. A more dynamic approach to ES assessment is to use spatially explicit simulations that represent the reciprocal interactions between the organisms providing ES and the environments they inhabit. Irrespective of the assessment approach adopted, the integration of stakeholders into the ES assessment provides an opportunity to engage those who make decisions about land-use in the process. In this thesis I aim to improve the understanding of ES relationships, the effects of land-use change on ES, and the use of alternative scenarios in the management process to sustain ES in multifunctional landscapes. I used the Rangatāiki Catchment in the Bay of Plenty region in New Zealand as a case study, as it is characteristic of other multifunctional landscapes in New Zealand. I took two approaches to ES assessment: (i) a ‘classic’ static mapping approach to ES and (ii) a dynamic, agent-based modelling approach. In the first, I analysed nine ES to identify areas in the catchment with ES supply high (hotspots) and low (coldspots), the relationships among ES, and where ES trade-offs and synergies occur. For the dynamic approach, I developed and tested (with detailed baseline and sensitivity analyses) a landscape-level pollination ES simulation model that estimates patch visitation rates by social nesting bees. In both approaches, I evaluated stakeholder developed landscape scenarios to understand the consequences of land-use change and landscape patterns on the spatial distribution of multiple ES. In my research I demonstrated how static ES mapping approaches and hotspot analysis illustrate the potential trade-offs in ES provision for potential future land-use scenarios. I also showed how scenario development and ES simulation modelling provide a ‘toolkit’ to promote a more participatory process in land-use policy and landscape design.