Spatially distributed climate data are essential inputs to the assessment and management of natural and human systems. The earliest applications of thin plate smoothing splines to the interpolation of surface climate were described by Wahba and Wendelberger (1980), Hutchinson et al. (1984ab) and Hutchinson and Bischof (1983). This methodology has been further developed into the ANUSPLIN software package at the Australian National University over the last 30 years. ANUSPLIN has become one of the leading methods in the development of spatial climate models and maps from ground-based data and has been applied in North America, China and many regions around the world (New et al. 2002; Hijmans et al. 2005; Hong et al. 2005; Rehfeldt 2006; Hutchinson et al. 2009; McKenney et al. 2011). The monthly mean climate surfaces produced by ANUSPLIN are fundamental components of the ANUCLIM climate application package that can be used to model species distributions in current and projected future climates (McKenney et al. 2007ab; Manning et al. 2010; Xu and Hutchinson 2013; Booth et al. 2014). Hydro-ecological applications and assessment of the nature and impacts of projected climate change are becoming major foci of this project.
The thin plate smoothing spline method is able to account for spatially varying dependences on process-based predictors, particularly elevation, a dominant predictor that is closely aligned with many controlling physical factors (Fig 1).
Annual mean precipitation over the coastal topography of northeast Queensland as mapped by ANUSPLIN.
The method permits robust and stable determination of dependencies on the predictor variables in remote data sparse, high elevation regions.
Explicit comparisons with kriging have been examined by Hutchinson (1993) and Hutchinson and Gessler (1994). Comparisons with other climate interpolation methods have been made by Price et al. (2000) and Jarvis and Stuart (2001). An extension to additive regression splines has been developed by Sharples and Hutchinson (2004). Scales of vertical and horizontal dependence of precipitation have been assessed by Hutchinson (1995a, 1998b), Sharples and Hutchinson (2005) and Johnson et al. (2016).
The method has been applied to the interpolation of Australia-wide monthly mean surfaces describing the principal climate variables that determine biological activity and hydrological processes. These include daily maximum and minimum temperature, precipitation, pan evaporation and solar radiation for the standard period 1976-2005. These surfaces can be used as baseline data for assessing potential impacts of projected climate change (Houser et al. 2004; Hutchinson and Xu 2015).
Monthly mean surfaces can also be used to underpin interpolation of climate variables over actual monthly and daily time steps. This has also become a major focus of the project with applications to Australia’s National Carbon Accounting System (NCAS) (Kesteven et al. 2004) and more recently to the development of fine scale grids of monthly and daily climate for Australia in collaboration with the University of Sydney and NSW Department of Primary Industries. These grids are being made generally available to the ecosystem research community as ANUClimate Version 2.0.
Applications of the spatial climate models are wide ranging. They have included assessments of agro-forestry for Natural Resources Canada (McKenney et al. 2011,2013; Pedlar et al., 2015), the impact of projected climate change on Australia’s snow conditions (Hennessy et al. 2008), the impact of drought on suicide (Hanigan et al. 2012), a widely used agroclimatic classification for Australia (Hutchinson et al. 1992,2005) (Fig 2), biodiversity in Australia (Phillips et al., 2013,2014; Gibson et al. in press) and Madagascar (Stalenberg et al. in press), continent-wide ecosystem assessment (Bloomfield et al. 2018) and food security in Malawi (Kawaye and Hutchinson 2018).
Fig 2 - Agroclimatic classes mapped across Australia by ANUSPLIN.
- Further refinement of methods for standardising monthly means of observed data to standard periods, to maximise the information obtained from stations with incomplete records, particularly in areas with a sparse data network.
- Further development of the techniques for interpolation of actual monthly and daily interpolation of precipitation and temperature across Australia. This is in further support of the National Carbon Accounting System (NCAS) and ANUClimate as well as providing baseline climate data for the assessment of climate change across a range of spatial and temporal scales.
- Assessment of climate extremes, their trends over time, and in projected future climates (Johnson et al. 2016; Crimp et al. 2018).