C ontemporary ecosystem change driven by a suite of global anthropogenic stressors has had reverberating consequences across genetic, population, community, and ecoregional scales (Díaz et al. 2019). Fine-scale changes in phenology, morphology, abundance, gene frequencies, and distribution of populations and species (eg Staudinger et al. 2013) can scale up to system-level conversions and biome shifts (Scheffer et al. 2009). Often driven by changing climate, many of these changes are manifest in ecological and physical stresses, including invasive-plant incursions, drought, desertification, severe fire, pest outbreaks, and geographic displacement of species. Extreme ecosystem changes are occurring with increasing frequency across a range of biomes, including coral bleaching in the tropics and grassification of shrublands (Figure 1). Ecosystem changes are expected to continue across many biomes even under scenarios with aggressive reductions in greenhouse-gas emissions, with globally distributed and radical ecosystem alterations predicted under high-emission scenarios (Nolan et al. 2018;Reid et al. 2018).We define these intensive and comprehensive system changes as ecosystem transformation (ie the emergence of a selforganizing, self-sustaining ecological or socioecological system that diverges considerably and irreversibly from prior historical ecosystem structure, composition, and function; Noss 1990). Transformations include ecosystem disruptions (eg Embrey et al. 2012) and occur across a range of temporal scales -for instance, from single-event high-intensity fires (Guiterman et al. 2018) to glacial-interglacial transitions spanning many millennia (Nolan et al. 2018) -and range widely in spatial extent, from a local community to entire biomes (Thompson et al. 2021). These changes pose critical threats to ecosystem services and consequently to human health and well-being, clean air and water, food security, sanitation, and disease mitigation (Whitmee et al. 2015).

Ecosystem transformation can be defined as the emergence of a self‐organizing, self‐sustaining, ecological or social–ecological system that deviates from prior ecosystem structure and function. These transformations are occurring across the globe; consequently, a static view of ecosystem processes is likely no longer sufficient for managing fish, wildlife, and other species. We present a framework that encompasses three strategies for fish and wildlife managers dealing with ecosystems vulnerable to transformation. Specifically, managers can resist change and strive to maintain existing ecosystem composition, structure, and function; accept transformation when it is not feasible to resist change or when changes are deemed socially acceptable; or direct change to a future ecosystem configuration that would yield desirable outcomes. Choice of a particular option likely hinges on anticipating future change, while also acknowledging that temporal and spatial scales, recent history and current state of the system, and magnitude of change can factor into the decision. This suite of management strategies can be implemented using a structured approach of learning and adapting as ecosystems change.

We used quantile regression to compare the body condition of walleye Sander vitreus and white bass Morone chrysops before (1980–1988) and after (1989–2004) the establishment of alewives Alosa pseudoharengus in Lake McConaughy, Nebraska. Higher quantiles (percentiles = 100% × quantiles (0, 1)) of weight (W) at the same total length (TL) were indicative of better body condition in an allometric growth model that included separate slopes and intercepts for the before and after groups. All quantiles of walleye weights by TL increased in the years after alewife introduction, ranging from 1.01 to 1.12 times weights in the years before alewife introduction, with greatest increases for the lower (<0.50) quantiles and greater TLs. Quantiles up to 0.25 (the lowest 25th percentiles) of white bass weights were reduced in years after alewife introduction for TLs less than 300 mm, ranging from 0.78 to 0.98 times weights in the years before alewife introduction. However, quantiles greater than or equal to 0.50 (the upper 50th percentiles) of white bass weights increased for all TLs, ranging from 1.01 to 1.06 times the pre‐1988 weights. A three‐group analysis, which improved the model fit for longer white bass, indicated a reduction (0.80–1.0) in white bass body condition across all TLs in the first 2 years (1989–1990) after alewife introduction, whereas body condition actually improved (1.02–1.12) across all TLs in later years (1991–2004). Thus, after the establishment of alewives in 1988, walleye body condition improved for all fish at all lengths (the greatest improvement occurring among fish in poorer condition), whereas white bass body condition was initially reduced for all fish at all lengths for 2 years and improved in subsequent years. The approach that we developed for comparing fish body condition before and after a management action in Lake McConaughy could be applied to other weight–length data sets typically evaluated with relative weight indices.

Intensifying global change is propelling many ecosystems toward irreversible transformations. Natural resource managers face the complex task of conserving these important resources under unprecedented conditions and expanding uncertainty. As once familiar ecological conditions disappear, traditional management approaches that assume the future will reflect the past are becoming increasingly untenable. In the present article, we place adaptive management within the resist–accept–direct (RAD) framework to assist informed risk taking for transforming ecosystems. This approach empowers managers to use familiar techniques associated with adaptive management in the unfamiliar territory of ecosystem transformation. By providing a common lexicon, it gives decision makers agency to revisit objectives, consider new system trajectories, and discuss RAD strategies in relation to current system state and direction of change. Operationalizing RAD adaptive management requires periodic review and update of management actions and objectives; monitoring, experimentation, and pilot studies; and bet hedging to better identify and tolerate associated risks.

No abstract