Ecological restoration in the age of apocalypse

Billions of dollars are spent annually on ecological restoration efforts around the world and yet successful attainment of restoration targets still falls short in many regions. Globally, ecosystem restoration is becoming increasingly challenged with changes in climate. Years with extreme climatic events that limit plant establishment, such as severe drought, heatwaves, and floods are projected to increase in frequency. A critical evaluation of current ecological restoration practices and changes to those practices are needed to attain global restoration targets. For plant restoration, many efforts globally focus on planting in a single year following disturbance. The odds of restoration efforts being conducted in a year that is inconducive to plant establishment may be calculated using climatic risk data. We propose a risk‐mitigation approach to restoration wherein plantings are conducted across multiple years for projects in a bet‐hedging strategy and evaluated through an adaptive management approach.


| INTRODUC TI ON
Increased likelihoods of extreme climatic events in a given year will impact ecological restoration efforts globally. Restoration of degraded terrestrial ecosystems is posited as a solution to climate change (IPCC, 2021; United Nations Environment Programme, 2021), but we now need to restore ecosystems in the midst of a rapidly changing climate with extreme annual and interannual variation. We propose a multi-year bet-hedging approach that considers increasingly variable abiotic and biotic conditions into account and becomes a standard for restoration practice on a broad global scale.

| CURRENT UNDER PER FOR MAN CE IN RE S TOR ATI ON DE S PITE EFFORTS
At present, many terrestrial plant-focused ecological restoration practitioners around the world actively direct seed or plant seedlings in a single year. A recent review of global dryland restoration efforts across six continents conducted in a single year estimated 17% of seeding projects failed to recruit even a single individual (Shackelford et al., 2021).
Many restoration projects around the world are planned following key drivers of degradation, some restoring plants over large areas (100s of 1000s of hectares). Often, organizations respond to such degradation F I G U R E 1 Multi-hazard resilience, with the heightened likelihood of extreme events such as droughts, heatwaves, wildfires, floods, and pest plagues, which drive ecosystem stress, creates an increasing need for adaptable restoration planning and design. Restoration scenario 1: The odds of planting limited seed sources in 1 year with the likelihood of an extreme event results in poor restoration success. Restoration scenario 2: A bet-hedging strategy to seeding, using seed over multiple years to overcome climate uncertainties, allowing for flexibility in the restoration design. Restoration scenario 3: Adaptive management strategy, through multi-year seeding plan as scenario 2, but with the inclusion of tailored seed enhancement technologies (e.g., seed coatings with surfactants or pelleting seed mixes) to ameliorate abiotic and biotic stressors, increasing the chance of seedling survival and restoration success (Brown et al., 2021), given the odds of survival are still very low in any given year. triggers by immediately applying all available resources to as much of the area impacted as possible utilizing only a few treatment and application methods over areas that are heterogeneous in both space and time. Seeding or planting projects installed in a single year is a high-risk management strategy as the odds of one or more extreme events that limit plant establishment occurring in that year are at historic highs and will likely increase into the future (IPCC, 2021) ( Figure 1, Restoration scenario 1).

| WE NEED TO RE THINK OUR APPROACH TO ECOLOG I C AL RE S TOR ATI ON
Landscape heterogeneity and uncertainty in environmental conditions have led researchers to explore bet-hedging approaches to ecological restoration through both space (Doherty & Zedler, 2015;Rader et al., 2022) and time (Davies et al., 2018). Bet-hedging in ecological restoration entails the utilization of multiple restoration treatments such that microsite conditions necessary for seedling survival in the initial stages of growth are available in at least some spatiotemporal contexts (Davies et al., 2018;Doherty & Zedler, 2015;Rader et al., 2022). In order to achieve the goals set out for the UN "Decade on Restoration," ecological restoration projects require scenario planning and adaptive management to maximize their success relative to costs, such as through a multiyear bet-hedging approach (e.g., Rader et al., 2022) which may be accomplished in four steps.

| S TEP 1: DE TERMINE THE RIS K S OF A PL ANTING YE AR HAVING AN E X TREME E VENT(S)
Projects need to model the abiotic and biotic variables that may impact seedling survival over an extended period to quantify the uncertainties and the likely performance of restoration designs as opposed to restoring species under assumptions of historic ranges of variability (e.g., Rader et al., 2022). Disturbances such as extreme climatic events, droughts, floods, and wildfires are increasing and shifting in intensity, frequency, extent, and location around the world (Seidl et al., 2017). For example, in certain regions globally, heavy precipitation events over land are projected to be 1.5 times higher with a 1.5°C global temperature rise, increasing the fraction of global land area affected by flood hazards (IPCC, 2021). A single event of this nature (e.g., a heatwave or flood) will likely have major negative impacts on seedling survival in and of their own (Shackelford et al., 2021). However, the likelihood of these climatic events being compounded by associated disturbances, like landslides, pathogen outbreaks, plagues, or wildfires, for a multi-hazard scenario, is high (Seidl et al., 2017) ( Figure 1).

| S TEP 2: UTILIZE RIS K DATA TO INFORM A B E T-HEDG ING S TR ATEGY
Even short-term exposure to extreme heat (minutes) can cause seedling mortality in many species (Teskey et al., 2015). Therefore, with projections of a three-time increase in heatwave events (IPCC, 2021), the risk of ecological restoration failure may increase by three-times not accounting for related drier conditions or other associated hazards (Seidl et al., 2017). Similarly, ecological and agricultural droughts are projected to be two-times more likely to occur in certain regions under the present 1°C increase (IPCC, 2021). Novel levels of drought alone are attributed with causing complete ecological restoration failure, even when intensive restoration methods such as outplantings that have historically demonstrated successful establishment are utilized (Rader et al., 2022).
Consecutive years of extreme climatic conditions, such as a year with a record drought followed by a year of heavy precipitation and flooding, both of which can lead to high seedling mortality, may also occur (Doherty & Zedler, 2015). In order to account for consecutive years with extreme climatic conditions that may cause seedling mortality multiple years of restoration planting may be needed to ensure at least some plants have a chance of surviving. As such, multi-year bet-hedging restoration plans that incorporate a feasibility study of regional climatic projections in which precious limited seed has a greater likelihood of success in at least one of a few years are needed (see Figure 1, Restoration scenario 2).

| S TEP 3: FUND MONITORING AND ADAP TIVE MANAG EMENT OF THE PROJEC T TO INFORM EFFORTS
Many abiotic and biotic barriers to native plant establishment must be overcome in order to succeed in restoration. As we still lack a clear understanding of the extent and timing of these barriers and the largescale impacts of extreme climatic events on recruitment (Shackelford et al., 2021), monitoring and adaptive management are required to inform future efforts. Seedlings can desiccate and die due to historically hot and dry summer conditions (Teskey et al., 2015) with extreme climatic events likely exacerbating barriers associated with restoration. The effects of extreme climatic events on established, intact systems are dire with major mortality of mature plants occurring (Seidl et al., 2017), but monitoring and research to further our understanding of specific barriers (Shackelford et al., 2021) and co-occurring barriers (Seidl et al., 2017) to plant survival at multiple growth stages is needed.

| S TEP 4: MITI G ATE RIS K S WITH AN ADAP TIVE MANAG EMENT APPROACH AND B E T-HEDG E YOUR TRE ATMENTS
Finally, applying multiple treatments such as microtopography alteration (Doherty & Zedler, 2015), vertical mulch (Rader et al., 2022) or innovative seed enhancement technologies (Davies et al., 2018) may further improve plant recruitment by increasing spatial heterogeneity of optimal microsite conditions. While physical alterations in microsites such as mounding and vertical mulch implementation can improve ecological restoration outcomes (Doherty & Zedler, 2015;Rader et al., 2022), these methods may be logistically unfeasible for restoration efforts with large-scale degradation like million hectare post-fire areas (e.g., Doherty & Zedler, 2015). Even in the absence of resources for barrier model-

| CHANG ING THE CLIMATE OF INVE S TOR MODEL S TO MEE T THE CHANG ING ENVIRONMENTAL CLIMATE
The increased risk of ecological restoration failure, especially for large-scale restoration efforts, not only poses an environmental problem, but also an economic one. For example, seed inputs can be expensive and costs can increase if establishment from seed is unsuccessful and ecosystem services are not restored (Shackelford et al., 2021). Thus, maximizing both the money spent and the resources expended on restoration efforts is critical for species conservation (United Nations Environment Programme, 2021). As we move through the next decade, we propose a greater focus of funding for on-the-ground efforts that utilize an adaptive, risk-mitigation strategy wherein limited seed resources are maximized. The number of years planned for seeding can be based on risk analyses, which can aid restoration managers to visualize the probability of success and assist in prioritizing areas for restoration.
Through multi-year seeding with diverse treatments, we increase the probability that favorable conditions are captured for plant establishment. Synergistic relationships between government agencies, non-governmental organizations, and private restoration endeavors are needed for the successful restoration of ecosystems globally. The Convention on Biological Diversity met this year to reset targets and goals that were not attained in the previous decade (United Nations Environment Programme, 2021), suggesting we need more proactive restoration efforts and thorough evaluation of current practices. While this article focuses on terrestrial plants, the same may be true for aquatic systems and animal species restoration as well. If we are to attain the new goals set forth by multiple international groups for restoration, critical evaluations of current practices and active changes need to be made.

AUTH O R CO NTR I B UTI O N S
LNS and ALR conceived and wrote the original draft of this manuscript, ALR created Figure 1, and LNS, ALR, and KWD edited and collaboratively discussed the development of this manuscript.

ACK N OWLED G M ENTS
We thank Drs Elizabeth Leger, Rory O'Connor, Tony Svejcar, Nancy Shackelford, Karen Holl, and three anonymous reviewers for comments and reviews to drafts of this manuscript. ALR is funded by the Australian Research Council, grant LP170100075. LNS is funded by the US Department of Agriculture-Agricultural Research Services post-doctoral position.

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors confirm they have no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing not applicable to this article as no datasets were generated or analysed during the current study.