Robust estimates of Natural Capital for GB
Peter Henrys Center for Ecology & Hydrology, Lancaster
Venue: Room A54, Postgraduate Statistics Centre, Lancaster University
Date: Thursday 29-09-2016, 4 - 5pm
Understanding the state of natural capital and the benefits to society that can be delivered is crucial for managing the environment in a sustainable way. Producing robust estimates of stock and change of natural capital indicators can, however, be complex and is heavily reliant on high quality data at sufficient spatial and temporal resolution. Understanding the spatial distribution of a natural capital asset and the pattern of change is critical to identify areas that are most at risk or that need preserving.
In this talk I will describe a GB-wide survey collecting multiple natural capital indicator metrics from coincident locations and a simple modelling approach that produces robust estimates of natural capital indicators. I will then describe how the methodology has changed to incorporate multiple sources of data resulting from a shift in emphasis to cheaper monitoring sources in light of resource constraints. Two examples will be presented: producing estimates of woodland extent across Britain based on data from satellite imagery and field survey methods; and also nationwide abundance of heather using data from field survey and ad hoc opportunistic citizen science records.
Finally, I will demonstrate an approach to mapping natural capital that maximises the use of information via a model based geostatistical approach implemented using INLA. This is compared to traditional approaches to mapping natural capital based on GIS-classification techniques.