The architecture of resilient landscapes: scenario modelling to reveal best practice design principles for climate adaptation

Adaptation Research Grants Program
Veronica Doerr
Year Started: 

Executive summary from final report:

Climate change is expected to result in significant changes in temperature, rainfall and evaporation, with the degree of change projected to accelerate within just a couple of decades.  As a result, Australia’s native species will experience quite different local environments than they do now and they will need to adjust to those environmental changes, move to live elsewhere, or go extinct.  Large populations and well connected natural areas may be required for species to make these adjustments, but both of these have been impacted by alteration of land uses and fragmentation of natural areas.  Thus, landscape design and management is one of the primary ways in which land managers can adapt their management of biodiversity under climate change.  Under landscape design and management, areas to be managed and/or restored for biodiversity are planned in very specific locations over relatively large scales with the aim of achieving more than a sum of the parts – not just lots of small populations but large populations that are spread over multiple patches of native ecosystems in the landscape, intermingled with other necessary land uses.

Many landscape design and management initiatives are underway in Australia and they differ substantially in their specific details.  Unfortunately, it is not yet clear whether one set of these ‘landscape design principles’ is better than another as a climate adaptation action.  This is in part because landscape design principles are developed and implemented based on analyses of current landscapes rather than future, climate-affected landscapes.  Yet future landscapes may be very different, not only in terms of where we might expect to find particular native species but also in terms of land uses including the types of crops grown, the extent of urban areas, and the amount of land devoted to relatively new land uses like carbon sequestration.  All these potential changes to landscapes could affect where native species live and the degree to which the landscape is connected to allow species movements.  Thus, we need to evaluate how well different landscape design principles perform in future landscapes.  And because we can’t predict exactly what future landscapes will look like, we need to consider a broad range of possible futures and try to identify landscape design approaches that are likely to assist native species across all of them.

To accomplish these goals, we modelled a range of plausible future landscapes and applied the most common current landscape design principles to these landscapes (as well as an aspirational design principle).  We then evaluated the degree to which the design principles might improve the capacity of the landscapes to support populations of native species in the long term, and decrease their capacity to support two key invasive species.  Our goal was to find one or more landscape design principles that improved all future landscapes for native species, as such an outcome would allow us to plan for the future without having to know precisely what the future will look like.

We worked with two case study landscapes: South-East New South Wales (which incorporated the Southern Rivers, Murray and Murrumbidgee Catchment Management Authority areas) and North-East New South Wales (which incorporated the Border Rivers/Gwydir, Namoi and Northern Rivers Catchment Management Authority areas).  We modelled 48 future versions of each of these landscapes based on:

  • Four ‘storylines’ of land-use change linked to different potential future climates as well as social and economic drivers and barriers of land-use change (defined with a group of experts across disciplines including agriculture and forestry)
  • Two global climate models which project differing rainfall patterns associated with each future climate and were used to model two versions of where native vegetation communities will be in the future
  • Six landscape design principles based on the amount and placement of new areas of native vegetation in the landscape (three determined based on structured interviews with Australian landscape managers about where they currently place vegetation management/restoration projects in their landscapes, one with random placement of new native vegetation, one with no new native vegetation, and one with a much greater area of native vegetation restoration than currently considered achievable)

We then evaluated our 96 future landscapes using a ‘metapopulation capacity’ model, which uses data on species’ habitat preferences and movement abilities to estimate a landscape’s ability to support populations that are large enough to persist long into the future.  We ran the model for four groups of native species (native orchids, animals that specialise on wet forest environments, and two groups of animals that specialise on grassy woodland and dry forest environments) as well as two invasive species (red fox and peppercorn tree).  We were hoping to find that one or more of the current, implementable design principles tended to do a better job of improving the metapopulation capacity of landscapes (increase it for native species groups and decrease it for invasive species) relative to the other principles across all the future landscapes.

Instead, we found that only our aspirational design principle – restoring landscapes to ~30% native vegetation cover – improved future landscapes relative to current landscapes.  None of the currently used design principles was better than another at improving the ability of a landscape to support native species in the long term, at least when assessed at these large landscape scales.  Even random placement of new areas of native vegetation achieved similar outcomes for biodiversity on average as principles based on careful spatial planning.  However, there were some differences between species.  The capacity to support wet forest specialist fauna declined regardless of design principles, and the invasive peppercorn tree increased with landscape improvements for native species.  Improvement in the capacity to support populations into the future also depended on land-use change storyline, so spatial planning of changes in land use may provide an additional management approach to climate adaptation for biodiversity management.    

Collectively, these results suggest that current approaches to landscape design and management may not be sufficient to serve as climate adaptation strategies for biodiversity.  The total amount of restoration is more important than detailed spatial configuration to counteract declines in biodiversity from climate-related changes in land use and suitable habitat, at least at very large landscape scales.  A variety of new approaches to landscape design thus need to be explored, including planning the placement of productive land uses in the landscape to achieve biodiversity co-benefits.  In the meantime, our results suggest that the best current no-regrets options include:

  • aligning local efforts over large scales but empowering local managers to design landscapes based on local knowledge and goals
  • developing complementary management strategies for a few types of invasive species that are most likely to benefit from restoration at landscape scales
  • concentrating effort to achieve ~30% native vegetation cover over smaller priority areas

Read the summary

Read the final report