Coastal marine ecosystems are predicted to respond to climate change through habitat loss, species range shifts, and biodiversity reductions. Distinguishing climate impacts from natural variability is challenging. Marine heatwaves (MHWs), influenced by global oceanic and atmospheric drivers, are increasing in frequency, duration, and intensity. In eastern boundary currents like the California Current, MHWs strongly correlate with temperature-nutrient dynamics, impacting primary productivity. Kelp forests, vital nearshore habitats, are particularly vulnerable. Northern California experienced unprecedented bull kelp (*Nereocystis luetkeana*) loss from 2014 to 2019, impacting communities, economies, and fisheries. This study investigates the role of a multi-year MHW and concurrent sea star wasting syndrome (SSWS) in causing this abrupt ecosystem shift, considering the historical resilience of the system and proposing adaptive management strategies.

Canopy-forming kelp species are foundation species in temperate rocky coastlines, providing crucial ecosystem services. Global kelp forest loss has increased in the last 20 years due to anthropogenic and climate-related factors. Ocean warming has been linked to kelp forest declines globally, including localized regions in Western Australia, the Tasman Sea, New Zealand, Baja California, Nova Scotia, and northern California. While the direct effects of MHWs on kelp forests aren't fully understood, they can alter ecosystem structure through changes in species composition, leading to shifts from healthy forests to algal turf reefs or sea urchin barrens. These shifts often involve cascading interactions across trophic levels, influenced by bottom-up (environmental) and top-down (predation) processes. The interplay of climate change and existing human impacts exacerbates the vulnerability of these habitat-forming species.

This study used a 34-year time series of kelp canopy coverage derived from USGS Landsat imagery. Multiple endmember spectral mixture analysis (MESMA) was used to generate this timeseries, offering a unique perspective on kelp's response to both acute climate events and in situ biological trends. The data were analyzed in conjunction with large-scale, local-scale, and biological drivers. Large-scale environmental forcings included the Multivariate El Niño/Southern Oscillation Index (MEI), North Pacific Gyre Oscillation (NPGO), and Pacific Decadal Oscillation (PDO). Local-scale forcings included sea surface temperature (SST), surface nitrate concentrations (NO3), and significant wave height (Hs). Biological indices (purple sea urchin (*Strongylocentrotus purpuratus*) and sunflower star (*Pycnopodia helianthoides*) densities) came from California Department of Fish and Wildlife (CDFW) and Reef Check California subtidal surveys. Partial least squares regression (PLSR) was used to analyze the temporal response of kelp canopy coverage to these drivers, accounting for collinearity among predictor variables. Ordinary least-squares regression (OLSR) was used to analyze changes in relevant biological and environmental indices before and after the MHW event.

Northern California bull kelp showed a dynamic interannual pattern of canopy coverage before 2014. The onset of the Northeast Pacific MHW and the mass mortality of sunflower stars (key urchin predators) coincided with dramatically reduced kelp canopy area in 2014. While anomalously cool, nutrient-rich conditions existed in 2012 and 2013, canopy area fell sharply in 2014 and remained suppressed despite more favorable conditions later. The spatial range of bull kelp also compressed. PLSR analysis revealed that including grazer dynamics in the predictive model improved accuracy in representing sustained low kelp biomass after the MHW. Environmental drivers (nitrate, SST, large-scale ocean-atmospheric forcing, wave height) alone predicted partial kelp recovery in 2017 once SST and NO3 rebounded, but incorporating urchin dynamics showed persistent low biomass despite environmental improvements. The 1997/1998 El Niño event caused kelp fluctuations but not widespread collapse, highlighting the importance of the sunflower star decline in reducing ecosystem resilience. Increases in purple urchin densities lagged sunflower star declines, linked to reduced predation and high larval recruitment. The stepwise decline in bull kelp wasn't observed in giant kelp, possibly due to differences in life history and physiological thresholds. Anomalously high sea urchin larval recruitment events did not always correlate with subsequent low kelp conditions. The historical decline in predator diversity lowered ecosystem resilience. The removal of sunflower stars removed existing system-wide resilience mechanisms. While temperature and nutrient conditions improved slightly in 2017, the kelp forest failed to recover, likely due to high urchin densities. The study suggests that additional mechanisms, such as disease outbreaks or human intervention, may be necessary to reduce urchin populations and allow a phase shift back to a forested state.

This study demonstrates that the dramatic decline of bull kelp forests in northern California resulted from a combination of factors: loss of top-down control on urchin populations, persistent shifts in SST and nutrient conditions beyond kelp's physiological thresholds, and a surge in purple urchin population and grazing intensity. The annual life cycle of bull kelp makes it especially vulnerable to acute stressors like MHWs and prolonged nutrient depletion. The differences in response between bull kelp and giant kelp underscore the importance of species-specific responses to climate change. The historical decline in predator diversity prior to the MHW suggests a reduction in ecosystem resilience, making the system more susceptible to the combined effects of the MHW and the sea star wasting event. The persistent high urchin densities, coupled with reduced kelp biomass, may represent a stable alternative ecosystem state, making recovery challenging. Despite a potential return of kelp under favorable environmental conditions, urchin herbivory significantly hinders full recovery.

The study highlights the critical role of both environmental and biological factors in shaping kelp forest dynamics. The combined impact of the MHW, sea star wasting syndrome, and resulting sea urchin increase created a persistent shift to an urchin barren state. While environmental conditions improved, the lack of predator control prevented kelp forest recovery. This research emphasizes the need for ecosystem-based management strategies, including monitoring, threshold-based responses, and targeted restoration efforts, to address the widespread loss of kelp forests and the associated ecological and socioeconomic consequences. Future research should focus on understanding the mechanisms driving urchin population dynamics and the potential for ecosystem recovery under different management scenarios.

The study's reliance on satellite imagery for kelp canopy assessment might not capture all aspects of kelp forest dynamics at finer spatial scales. The limited availability of historical biological data (urchin densities before 2003) restricts the analysis of long-term trends and limits the ability to fully assess the role of historical predator dynamics in shaping ecosystem resilience. The study focuses primarily on the northern California region, and the findings may not be directly generalizable to other kelp forest ecosystems with different species composition or environmental conditions.