The California Hotspots Project: I.Identifying regions of rapid diversificationof mammalsEd Davis, M. Koo, C. Conroy, J. Patton & C. MoritzMuseum of Vertebrate Zoology, UC Berkeley*Funded by Resources Law Group & CA State Parks*Thanks to J. Grinnell & colleagues, T. Smith, V. Sork, D.Ackerly, T. Barnosky, D. Wake, A. Vandergast, A. Bohonak.Goal: To maintain evolutionaryprocesses and the viability ofspecies and functional landscapesnecessary to achieve this.Moritz 2002 (After Frankel (1974))A policy-driven challenge to evolutionarybiologistsActive process of landacquisition forconservation in CA:CSP, TNC & others:• $0.5B last 5 yrs for purchase• Prop 84: $6B 2007-2013• Get it right, make it count..• Include Evol Processes as amajor factor in planning toassist resilience• Rayburn - take a bow!Where are the areas ofactive diversification:Evolutionary hotspots?Environmental surrogates for evolutionaryprocesses: South Africa (Cowling et al.)Diversification across lowland-upland gradients,edaphic gradients, macroclimatic gradientsNeo-endemism ofCA plants(polyploid derivatives)Stebbins & Major 1965Hypothesis: Elevated neo-endemism in regions withrecent (Plio-Pleistocene)development of fullMediterranean climateAND not subject to recentglaciation(also Raven & Axelrod 1978)Endemism-richness for species vslineages (Rissler et al. 2006)Key evolutionary processesAdaptive diversificationAbiotic gradients(climate, soils)Biotic interactions ->character displacement,(eg. novel communities, suturezones)Genetic driftGenerative processes OpportunityGenetic isolationNiche spaceStabilityNon-adaptive diversificationWhat biological data do wehave/need to map Evol hotspots?• Geographic distributions & phylogeny (species,phylogeographic lineages)– Hotspots for short-branch taxa [neo-endemism]– Hotspots of phylogeographic endemism– Suture zones• Spatial information on ecotypic differentiation(phenotypes)– Subspecies endemism as a surrogateToday: mammals onlySubspecies as surrogates for geographicpatterns of phenotypic (adaptive?) divergence• Skepticism from somesystematists– May not reflect historical lineages(good)– May be artefacts of inadequatesampling/analysis of clines– Variable criteria for recognition (bad)• BUT, evolutionists (eg. Grinnell,Mayr) have long used subspecies todescribe geographic patterns ofphenotypic variation and viewsubspecies as intermediates in thespeciation processDiverse CA taxa:Dipodomys (30)Tamias (16)Thomomys (16)Microtus (12)Chaetodipus (12)General approachSelect (near)endemic taxa(>75% range in CA)26 species130 subspeciesEstimate ranges @ 1km resolution(MVZ points -> range-map-clipped MaxENT models)Richness mapsEndemism-richness = ! "1/area) per cellmtDNA distanceto sister species(GenBANK,unpubl)Species neo-endemism richness= !"(1/area) X 1/distance) per cellEndemic species - richness, endemism-richness & neo-endemism1. SF Bay area: R. raviventris2. Interior Coast Ranges: D. ingens, D.nitratoides, A. nelsoni3. Tehachapi: P. alticola4. Central Sierra: S. lyelli, T. alpinus12343,4 = recent lineages cf.1,2 = old;Note: north/central coast!"1/area) !"1/area)(1/distance)Note: islandendemics excludedEndemic species: foci of recent vs old taxa (not1/area weighted)Neo-Richness -! (1/distance)Paleo-Richness! distanceRichnessCentral coast,Tehachapi, CentralSierra are fociEssentially nodifference torichness; ie randomRichness & endemism-richness of endemic subspeciesNorth coast, SF Bay, Tehachapi, SanBernadino, CV, Owens Valley, CentralcoastNeo- and subspeciesendemism (top 10%) &comparison with CA/FedProtected AreasSpecies & subspecies neo-endemism areas distinct exceptTehachapis, St. Lucia, SanBernadino RangesEnvironmental correlates:coastal gradients,biogeographic overlap areas(gradients)Good potential for integratedreserve design in most areas(except Nth coast?).Note - other taxa & approachesneeded (eg. Sork,Vandergast/Bohonak etc.)MVZGrinnellProject:change over100 yrsAcross Yosemitetransect, many smallmammal showupwards range shiftsof 500-800m since1920 (with c. 3C increasein monthly minimumtemperature)Management & policy implications• Specific - don’t get excited yet - moretaxa/approaches to come!• General:– Rates of diversification are not randomly distributed- cold & hotspots– Common theme is gradients/ecotones, stability?– Location of hotspots likely to shift with climate– Protection of EHs requires large-scale andcoordinated planning to protect connectivity acrossgradients, and large, heterogeneous isolatesCaveats• Spatial errors in distribution estimates– Minimized by combining models & point estimates• Error in inferring relative divergence time frommtDNA distance– ancestral coalescence, introgression etc.) - likelyoverestimating divergence times; adds noise• Vagaries of subspecies classifications– But most from Grinnell (or his students) so relativelyconsistent approach• Current patterns of neo-endemism may reflectenvironmental, but not geographic space oforigin (eg. late Q range shifts)Components of Hotspots ProjectConceptualmodel: => environmentalpredictors; e.gtopographic, climaticand geologicalcomplexity, ecologicalstabilityEvolutionarytheory:Selection,isolation,hybridization etc.Spatial prediction ofrates of evolution=> hotspotsSpatial layers forenvironmentalpredictorsSpatial patterns ofdiversity (genetic,phenotypic, phylogenetic,species):Mammals, amphibian,reptiles, birds, plantsConservation values;gap analysis to identifypriority areas foracquisitionBuild &