Certain restoration objectives and targets will be more sensitive to climate change than others. For example, whilst recreational benefits can be achieved by a range of different landscape configurations, the recovery of a threatened species may be dependent on a narrow range of habitat conditions. This is illustrated at the Mayesbrook Climate Change Park, where 45 hectares were restored with the objective of helping a community adapt to flooding risk, exemplifying how urban greenspace could provide key ecosystem services (Restoring Europe’s Rivers, 2019). As the location centred around a large urban park, the ecosystem service and recreational benefits could be realised without specifying a target habitat type or targeted species reintroductions.

Restoration approaches often use historic conditions to set objectives and measure success. However, uncertainties in the characteristics of future, novel ecosystems (Corlett, 2016; Nolan et al., 2018) mean objectives may instead need to be adaptive and focussed on the rehabilitation and resilience of specific ecosystem functions (i.e. ‘ecological renovation’; Prober et al., 2019). Lessons can be learned from rewilding, which typically focuses on functionality rather than prescribed structural/compositional outcomes. There is also a need to consider novel ecosystem configurations as possible restoration targets. In relation to coral reefs, some sites are becoming non-coral systems, whilst coral species range shifts are creating new species configurations, interactions and functionality at other, sometimes previously non-coral sites (Graham et al. 2014).

The suitability of sites for restoration action will change as environmental conditions alter, species distributions shift and the intensity and footprint of pressures change. When identifying potential sites, it is important to consider the anticipated climatic changes in that area. This includes the response of species and ecosystem components (see Section “Account for future distribution and fitness when choosing target species and ecosystems”), the vulnerability and exposure to threats, and the space available for the system to expand or move. Furthermore, in order to increase the probability of achieving positive outcomes, reducing the drivers of habitat degradation and loss first and foremost is imperative before commencing restoration actions.

Using models of climate, vegetation, land use and biodiversity it is possible to refine areas suitable for restoration. For example, in the humid tropics, candidate areas for restoration have been identified by combining models that describe how plant functional types reorganise under climate change with spatial predictions of deforestation (Asner et al., 2010). This helped identify areas where high deforestation coincided with climatically more suitable conditions and climate refugia. In the Brazilian Atlantic Forest, candidate restoration sites were selected based on modelling that not only accounted for climate change and land use change but also considered socio-economic factors and management approaches such as protected areas (Zwiener et al., 2017). In selecting candidate sites it is also important to think about the wider land-/seascape, to facilitate and aid restoration by connecting patches (Tarabelli et al., 2005). However, in a climate change context, connectivity also needs to be managed in relation to potential negative impacts, for example where landscape corridors can aid the spread of invasive species (Resasco et al, 2014).

When determining which ecosystem components to restore, projected changes in environmental conditions and species fitness and distributions must be accounted for. Species distribution modelling (SDM) based on climate envelopes can help shortlist possible species to use in restoration projects. In identifying semiarid shrubland species of potential use for restoration in Central Mexico, Gelviz-Gelvez et al. (2015) analysed 46 species to select those with the most desired characteristics of sociability, cover, abundance, and micorrhization, and subsequently used SDM to further refine the list.

The projected impacts of climate change will influence whether natural regeneration (e.g. passive restoration involving locally adapted species) is preferential to active restoration using species within or outside their native ranges. In the latter case, taxon substitution may be necessary to replace a native species vulnerable to climate change with a non-native species able to fulfil the same or similar ecosystem function (Corlett, 2016). Experience in the latter option derives from the practice of assisted migration in response to climate change (Williams and Kasten Dumroese, 2013), where species at lower latitudes are predicted to fare better under non-analogous future climate conditions at higher latitudes, or indeed altitudes. However, this option is fraught with uncertainties because of cryptic maladaptation, underestimation of climate variability differences, and many unforeseen issues from introduced and/or invasive species (Benito-Garzón et al., 2013).

Local material is often the preferred choice in order to ensure good adaption to the local environment, maintain the genetic integrity of the site and prevent potential pollution of the gene pool (Harris et al., 2006). However, in the context of a rapidly changing environment this should be challenged as local provenance may be poorly adapted to future climates and disturbance regimes (Prober et al., 2015). Instead, material may be sourced from areas currently experiencing conditions similar to the predicted climatic conditions for the restoration site (Booth, 2016). At the same time, resilience and adaptability can be enhanced by genetic diversity within and among species, capitalising on the capacity of species to adapt to environmental change through plasticity, selection, or gene flow. In this way, space for evolutionary development can and indeed should be encouraged by restoration projects (Rice et al., 2003).

Prober et al (2015) call for a “climate-adjusted” provenance strategy, combining genetic diversity and adaptability, targeting the direction of projected climate change, whilst also allowing for uncertainties in those projections and in the possibility of unforeseen selective agents. In the case of coral reefs, climate change adapted coral material can be identified from colonies that have survived high temperature events and/or mass coral bleaching, representing promising genetic/epigenetic variants for reef restoration initiatives (Morikawa and Palumbi, 2019; Rinkevich, 2015). This was the case for the World Bank Coral Reef Targeted Research Program, which investigated coral reef gardening as a novel, cheaper strategy for restoring damaged coral reefs. Species selection was made climate resilient, since coral species that showed resilience to increasing sea surface temperature were selected for propagation (Shaish et al., 2008).

The biophysical environment may need modification to maintain or create favourable conditions for the ecosystem components and their desired outcomes. Recovery may suffer if ecosystem components are vulnerable to climate change, for example, if key micro-climatic niches or vital ecological functions are lost. Understanding how climate change will impact ecosystem interactions and climatic niches can lead to the development of efficient and targeted physical modifications to the ecosystem as part of the restoration project.

Restoration can involve physically changing the environment, such as installing dams to alter water levels and recreate the conditions suited to the restoration target. Climate change may add a new level of necessity to such environmental manipulation. For example, human-built structures may be needed to limit climate change impacts (e.g. sea water inundation) or give natural systems more space or time to adapt, by creating small-scale climate refugia. In south-western Australia, dense replanting along rivers has been proposed to keep water temperatures within the thermal tolerance of temperature-sensitive aquatic fauna (Davies, 2010), with a 10% increase in riparian cover generally leading to a reduction of water temperature by 1 °C. In the case of river floodplains, restoring connectivity can be important to reverse the effects of past river and floodplain engineering such as levees, ditches and channelization, and counter future reduction of flow levels and flood events (Perry et al., 2015).

Climatic changes have the potential to significantly impact the functioning of ecosystems. For example, phenological changes may limit the spatiotemporal availability of food sources, as already documented for some bird species (Carey, 2009; Mastrantonis et al., 2019). Hence, restoration initiatives may need to consider when and where the loss of critical species interactions and ecosystem functions may occur in the future. Conceivably, ecosystem engineering approaches, including species introductions, may be required to fill the roles of keystone species extirpated by climate change impacts, including extreme events such as floods and wildfires. The Vietnamese Coastal Wetlands Protection and Development Project (CWPDP) utilised both physical manipulation and species selection when restoring mangrove forests in southern Vietnam: plantations were specifically designed to be species-rich to provide genetic variety, following the lack of success with monoculture plantations in the north of Vietnam (Powel et al., 2011). Building dykes and bamboo fences also allowed for the natural rehabilitation of mangroves, whilst acknowledging that climate change and past land use had resulted in differing abiotic conditions that hindered natural regeneration (World Bank, 2008).

Many threats to natural environments are directly related to climate change, such as more intense and frequent extreme weather events and increasing sea surface temperatures. Other threats are the result of tangible human actions such as increased nutrient input leading to eutrophication, and some are the result of directed actions which are then exacerbated further by climate change, such as invasive species and the spread of disease.

These global change hazards can have local, site specific impacts and must be considered in order to enhance the resilience of restoration projects. Whilst certain risks may be minimised using the approaches outlined elsewhere in this framework, some might be difficult to abate. Potential impacts should be anticipated and mitigated within the design and implementation of the restoration activity. Climate change related risks are summarised in Fig. 2 and discussed in more detail below.

Fig. 2.

Hazards that have the potential to impact local restoration projects can be grouped into distinct themes. The ones shown here are not exhaustive.

(0.21MB).

Policies, land management options and stakeholder dynamics may be affected by climate change in ways which influence the feasibility and direction of restoration. This is exemplified in the Berau Forest Carbon Program, which aims to preserve and restore the natural native forest cover in Indonesia by implementing a REDD+ scheme and coordinating spatial planning and community empowerment. By aligning the project with Indonesia’s Low Carbon Growth Strategy scheme and becoming a recognised REDD+ activity, the program was able to benefit from guidance and cooperation from these policies (IGES Forest Conservation Team, 2013). In addition, the program attracted support from The Nature Conservancy, who aided in the remediation of tenure conflicts (Anandi et al., 2014).

Climate change policies, and other policies that can be affected by climate change, may influence the viability, design, implementation and ultimate success of restoration projects. Restoration may be incentivised by climate change policies and mechanisms (e.g. REDD+), but in some cases discouraged as when increased demand for biofuels raises competition for land (Brodie et al., 2012). Policy impacts may be direct, for example when a hydropower plant as part of a low carbon energy plan causes direct loss of a landscape undergoing restoration. They may also be indirect, for example when changes in agricultural suitability can create but also close opportunities for restoration (Bradley et al., 2012). Stakeholder and land management is therefore key to restoration success in locations other than officially protected areas. Adoption of secure conservation measures, such as conservation covenants with private landholders, purchase of strategically critical properties and expansion of the public protected area estate are being utilised in the Gondwana Link project in Australia, to reduce the predicted impact of increased land use competition under future climate change. The Gondwana Link project aims to provide 1,000 km of continuous habitat, from dry woodlands in the interior to tall wet forests in the south-west corner of Western Australia. This project also aims to build livelihoods and support traditional owners to manage the land (Bradby et al., 2016; Figgis et al., 2012; Jonson, 2010). Restoration projects such as this, designed around the ‘triple wins’ of joint mitigation, adaptation and development goals may be most resilient to policy influences (Favretto et al., 2018).

Policies and land management options can also be enacted at local levels within countries. The Atlantic Forest Restoration Pact is a collection of the efforts of many smaller organisations to restore parts of the Atlantic Forest in Brazil. The Pact has a covenant in which signatories define their priorities including sharing knowledge on forest restoration, providing advice to align restoration actions with the Forest Code (a law requiring landowners to maintain part of their property as native vegetation), and contributing to public policy formation to enhance outcomes. Through this Pact, smaller local restoration projects create greater impact by aligning their actions. Project proponents also benefit from knowledge and technology sharing, training opportunities and expansion of job opportunities that participation in the Pact brings (Pacto Mata Atlantica, 2016).

In a rapidly changing climate, restoration management decisions may be valid for only short time periods. For this reason, monitoring and adaptive management of restoration projects is key to ensuring their success (Jandl et al., 2019). This requires clearly stated and attainable objectives supported by appropriate indicators, sufficient monitoring resources and implementation plans. At the Stewart Lake Waterfowl Management Area in north-eastern Utah, restoration work aimed to reduce hazardous selenium contamination by careful water inflow/outflow control. This was sensitive to decreased river flow and increased risk of drought as predicted by climate change models (Farag et al., 2017). The management was designed to be adaptive. Data collected from monitoring water, sediment, plankton and fish selenium levels were used at regular stakeholder meetings to evaluate and refine the restoration strategy, informing water management and other measures going forward.

Adaptive management that allows for changes to restoration actions throughout the project is also important in aiding their resilience. For example, the Great Eastern Ranges (GER) project in Australia aims to work with communities to restore healthy, biodiverse forests, woodlands, soils and wetlands to restore carbon stocks (Dunn et al., 2012; Mackey et al., 2010a, 2010b). The project is working to link habitats across 3,600 km, from western Victoria through NSW and the ACT to far north Queensland. Research was conducted to identify and characterise the integrity of landscapes within the GER corridor. This information was used to understand patterns of land disturbance and the likely influence of climate change on future weed outbreaks. As a result, emergent outbreaks could be prioritised for immediate treatment before they take hold in a landscape, thus prioritising actions within an organised and proactive management framework.

Box 1.

Reforestation in Pontal do Paranapanema.

Overview

The Brazilian Atlantic Forest is a world biodiversity hotspot (dos Santos et al., 2019). Since 2002, the Institute of Ecological Research (IPE), Sao Paulo University and WeForest have worked to reforest Pontal do Paranapanema and create an ecological corridor to Morro do Diabo State Park. The original 10-year ‘Atlantic Corridors Project’ (AFCP) restored 890 hectares (WeForest, 2019), conserving biodiversity, promoting economic development, and building livelihood resilience. Following this success, WeForest joined the Atlantic Forest Restoration Pact, which intends to restore 15 million hectares of Atlantic Forest in Brazil by 2050 (Pinto et al., 2014). WeForest shares research findings to prioritise future areas for restoration (Scarano and Ceotto, 2015).

Areas where climate change was considered

Less than 7% of the original forest remains, and is highly fragmented (Tarabelli et al., 2005), so buffer regions that improve connectivity and facilitate climate-related distributional shifts are favoured. AFCP and IPE successfully lobby for the creation of buffer zones around forest fragments to facilitate species migration (2).

1,779,500 trees from community-based nurseries propagating 115 indigenous tree species were planted in designated land fragments (WeForest, 2019). Many have wide ranges and are fast-growing, acting as “pioneer” species in agroforestry practices and helping ensure the long-term resilience of the project (3).

The role of forests in regulating water supplies, and ability of restoration to re-establish ecosystem services, was explained in educational outreach. WeForest estimated that over 2,000 hectares of land are benefitting from improved water cycling and quality, helping to adapt to climate-related impacts of droughts and flooding (5).

The AFCP aimed to link land management and climate related policies by training landowners in sustainable agriculture and wider environmental education, with the aim of achieving longevity of the project even if the restored land comes under increasing pressure (6).

Conclusion

Additional strengths of the AFCP include the range of sites and techniques used, the range of partners, and the inclusion of socioeconomic criteria in monitoring protocols. There has been measurable success, including continuation of tree nurseries beyond project completion, and reports of fewer intentionally set fires. This is significant because fire can preferentially occur in regenerated secondary forest due to higher sun exposure; without adequate management, restoration success can be compromised (dos Santos et al., 2019) (Fig. I).

Fig. I.

Restored wildlife corridor in Pontal do Paranapanema. (WeForest).

(0.5MB).

Box 2.

Crayweed restoration in New South Wales

Project overview

In 2011 the Sydney Institute of Marine Science commenced crayweed forest restoration in Sydney. This macroalga once formed extensive underwater forests providing habitat for a diverse fauna including species important for wild caught fisheries. ‘Operation Crayweed’ aims to implement large-scale restoration, increase community engagement, and improve awareness about the importance of seaweed forests to marine ecosystems (Vergés et al., 2020).

Areas where climate change was considered

A principal objective was ecosystem service re-establishment and prevention of further declines due to climate change. A strong awareness and education programme highlights to the public the importance of underwater forests to ecosystem functioning, through a combination of art, narrative storytelling and science. Upscaling Operation Crayweed to the entire Sydney Metropolitan region will meet most success if awareness raising is continued (Vergés et al., 2020) (1).

Genetic testing of crayweed indicated that populations are distributed spatially across the NSW coastline and plants could be sourced from a 60 km radius for restoration. Restoration success was measured by genotyping the first generation of crayweed, which exhibited patterns of genetic diversity and structure similar to donor plants and natural populations (Wood et al., 2020). Ongoing research looks for potential warm-adapted genotypes in crayweed to ‘future-proof’ restoration (3).

A Before After Control Impact design was implemented to monitor survival, recruitment, range expansion and epifauna assemblage under changing environmental conditions (Campbell et al., 2014). Early monitoring demonstrates that crayweed survival was high (~70%) and mobile epifauna assemblages on restored crayweed did not fully resemble those from reference crayweed populations (Marzinelli et al., 2015). This highlights that restoration of seaweed ecosystem assemblages is complex and a long-term process (7).

Conclusion

Over multiple generations, Crayweed range has expanded several hundred metres from original restoration sites and moved nearshore with minimal maintenance or additional plantings required (Vergés et al., 2020). The potential impact of rising sea surface temperature on herbivory is one area the project needs to address. Planning and management for novel herbivory under climate change will be crucial. This may include options such as determining optimum patch size and density for recruitment success, deterring herbivory (Westermeier et al., 2013) and restoring macrophytes when herbivory is lowest (typically in winter; Carney et al. 2005) (Fig. I).

Fig. I.

Juvenile crayweed being measured (John Turnbull).

(0.83MB).