Award: OCE-1233288

Undersea forests of giant kelp (Macrocystis pyrifera) are highly productive marine ecosystems that occur on shallow discontinuous reefs at the interface of the land-sea margin in temperate regions throughout the world. Not only are giant kelp forests ecologically important for sustaining a diverse array of plants and animals in the areas in which they occur, but they also provide valuable living resources, benefit the environment and have cultural value to a broad sector of society. A defining feature of giant kelp forests is that they are extremely dynamic and fluctuate greatly over time and space. Disturbance from storm waves, episodic grazing by sea urchins, and sporadic events of adverse growing conditions lead to frequent local extinctions that occur out of phase across the species? range. Local extinctions tend to be short lived as rapid recolonization by giant kelp following disturbance is the norm. The ability of giant kelp to rapidly colonize deforested reefs has puzzled researchers for years because of the perception that its propagules (microscopic spores) have a limited capacity to disperse across the distances required to repopulate neighboring reefs. Results from our prior NSF funded research disprove this perception by showing that kelp spores are indeed capable of dispersing the distances needed to colonize neighboring areas ("patches"). What has remained unanswered is the extent to which the dynamics of the aggregate population of giant kelp in a region (termed "metapopulation") are influenced by the degree to which its member patches are connected through spore dispersal. In this project we used a combination of satellite imagery of giant kelp, models of ocean circulation, molecular genetics, ecology and kelp natural history to determine the processes that cause fluctuations in giant kelp metapopulations throughout its range in California. We focused our efforts on measuring how local patches of giant kelp are connected to one another through spore dispersal and the physical and biological processes that influence this connectivity. Scientists have had difficulty testing how connectivity among member patches affects the dynamics of metapopulations in the natural world because this requires comprehensive information on the size and reproduction of all patches in a metapopulation, and on the amount of dispersal among them. There are very few systems for which this information exists. We were able to compile this information for giant kelp forests throughout its 2200 km range extending from central California to Baja California, Mexico by developing a 33-year (1984¬–2016) time series of giant kelp biomass developed from Landsat satellite imagery, which we combined with estimates of spore dispersal obtained from a high resolution ocean circulation model. We discovered that the degree to which a kelp patch is connected to other patches in the metapopulation strongly predicted local dynamics across the entire region. Well-connected patches had lower risks of extinction and higher chances of colonization, leading to greater likelihoods of giant kelp persistence. Moreover, we learned that the amount of connectivity between patches is governed more by fluctuations in the production of spores than by fluctuations in the currents that transport them. Interestingly, we found that individuals in isolated patches produced fewer spores than individuals in well-connected patches. Results of genetic analyses of these individuals were consistent with expectations of inbreeding, which we previously showed leads to reduced spore production. Genetic analyses of individuals collected from populations ranging from Alaska in the north to Baja California, Mexico in the south revealed five main geographic clusters, three of which were located in southern California. We found that connectivity by ocean currents explained the genetic structure of populations within these clusters suggesting a network of three giant kelp metapopulations in southern California. Breaks between clusters were not completely explained by oceanographic circulation models, but instead corresponded with previously identified biogeographic breaks. This suggests that environmental factors play an important role in the genetic isolation between adjacent clusters. The intellectual and broader impacts resulting from this award are varied and substantial. The findings validate theoretical predictions in ecology and population genetics and advance our understanding of these disciplines. Moreover, the results are broadly applicable to conservation biology and are useful for informing resource managers on policy issues pertaining to marine spatial planning and restoration. At the time of this report, this award resulted in nine fully documented data sets, contributed to 12 peer reviewed publications, 2 PhD dissertations, 1 Master?s thesis, and provided research training to 3 post docs, 6 graduate students and approximately 22 undergraduate students. Last Modified: 12/29/2017 Submitted by: Peter T Raimondi