Use of Assisted Migration and Community Zonation Patterns to Build a Climate- Resilient Coastal Landscape
Loretta L. Battaglia Hannah J. Kalk
Southern Illinois University Carbondale, Illinois USA Coastal plant communities and altered inundation regimes
Adapt, go extinct, or migrate inland
Credit: wetlandextension.ifas.ufl.edu With only 1 m rise in sea-level, 26-66% of coastal wetlands will be lost (Mitsch and Gosselink, 2007) Climate Change and Restoration
• “Moving targets” – Rate of change ExponentialFast Linear
– Direction of change Slow
Environmental Variable Environmental
Environmental Variable Environmental • Anticipatory or futuristic restoration Time
• Assisted migration
Relative Ease of Futuristic Restoration
Rate of Change
Slow Fast
Linear High Moderate Pattern of Spatial Exponential Moderate Low Change Erratic or ? ?? No-analogue Coastal Wetlands
• Spatial zonation of communities along strong environmental gradients
• Stress-tolerant species at seaward end, stronger competitors farther inland (Bertness and Ellison 1987, Pennings et al. 1995) Loretta Battaglia
• Rate of change varies depending on relative SLR, etc.
• N. Atlantic, Gulf of Mexico, especially at risk (Hammer-Klose and Thieler 2001, Meehl et al. 2005, Ramstorf 2007)
National Geographic Relative Ease of Futuristic Restoration
Rate of Change
Slow Fast
Linear High Moderate Direction of Spatial Non-linear Moderate Low Change Erratic or ? ?? No-analogue Study Questions
• With removal of biotic filters, can dominant species from seaward communities establish and survive when introduced into landward locations? (Reciprocal planting study – Weeks Bay, AL)
• Can assisted colonization enable successful futuristic restoration and if so, how futuristic? (FEMA Restoration Sites – Grand Bay, AL)
Gulf of Mexico
Northern Gulf of Mexico
• Relatively slow relative SLR
Grand Bay Weeks Bay NERR NERR
Gulf of Mexico Dominants of Seaward Vegetation Zones
• Salt marsh: Spartina alterniflora • Brackish marsh: Juncus roemerianus • Fresh marsh: Cladium jamaicense
Weeks Bay NERR
Forest- Terrestrial Forest Open Water Salt Marsh Brackish Fresh Marsh Marsh Marsh Ecotone Gulf of Mexico Methods
• Late Summer 2008 vegetation removed in 1m2 plots
• Five culms of 3 dominant species from 3 most seaward vegetation zones planted in monoculture
• Transplant zones (3 sp x 5 reps; total plots = 75) – Salt marsh (SM) seaward – Brackish marsh (BM) – Fresh marsh (FM) Weeks Bay NERR – Fresh marsh – Forest ecotone (EC) – Wetland seep forest (FO) landward
• Survival (and growth) monitored seasonally for 2 years
Removal of Standing Vegetation
Weeks Bay NERR
Gulf of Mexico Planting of Culms
Weeks Bay NERR
Gulf of Mexico Results 1.0 1.0 Spartina Juncus 0.8 0.8 a
0.6 0.6
0.4 0.4
Percent Survival Percent
Percent Survival Percent
ab 0.2 0.2
b b b 0.0 0.0 SM BM FM EC FO SM BM FM EC FO
1.0 Cladium 0.8
0.6 Weeks Bay NERR
0.4
Percent Survival Percent
0.2 Gulf of Mexico
0.0 SM BM FM EC FO Summary of Results
• All species established and survived in at least one of the zones landward of their original zone
• Juncus exhibited the broadest “new” distribution – Established seaward of original position – Thriving in forest (canopy openings from Hurricane Katrina in 2005) Weeks Bay NERR • Cladium was able to establish landward as well, but not in more seaward zones Gulf of Mexico
FEMA Buyout Homestead Sites
• 15 former pine forests adjacent to GBNERR
• Purchases began following inundation by Hurricanes Georges in 1998
• Highly disturbed Credit: L. Battaglia • Heavily infested with noxious exotics: Imperata cylindrica , Panicum repens and Triadica sebifera
• Vacant and awaiting management Credit: MDMR
Credit: L. Battaglia Credit: L. Battaglia Design of Propagule Bank Experiment Randomly Selected: 5 sites, 9m X 13m plots, 24- 1m X 1m subplots
Site Preparation: herbicide, mowing, tilling, raking
Legend S B F M W C
S: Salt marsh B: Brackish marsh F: Freshwater marsh M F C C B B M: Maritime pine island W: Wet pine flatwood C: Control F M W S W C
S M W B F W
Donor Bank Collection and Application 0.50m x 0.25 m sods
Data Collection and Site Maintenance
•Presence/Absence data
•Species removed: Eupatorium capillifolium, Centrosema virginiana, Ipomoea quamoclit and Cuphea glutinosa Richness and Diversity
• Site preparation greatly increased species richness and diversity • Highest in plots with freshwater sods, relative to plots with control and saline marsh sods
303.5
a a 253.0 b ab ab b b ab
202.5 ab
2.0 b 15 b b 1.5
Mean species richness species Mean 10 1.0
Mean species diversity (H') diversity species Mean 5 0.5
0 0.0 U C S B F MP PF U C PropaguleS sodB treatment F MP PF Propagule sod treatment Compositional Trends Site 1: Brackish Marsh Site 1: Control Site 1: Freshwater Marsh Site 1: Maritime Pine Site 1: Pine Savanna Site 1: Salt Marsh Site 2: Brackish Marsh Site 2: Control Site 2: Freshwater Marsh Site 2: Maritime Pine Site 2: Pine Savanna Site 2: Salt Marsh Site 3: Brackish Marsh Site 3: Control Site 3: Freshwater Marsh Site 3: Maritime Pine Site 3: Pine Savanna Site 3: Salt Marsh
Axis 3 Axis Site 4: Brackish Marsh Site 4: Control Site 4: Freshwater Marsh Site 4: Maritime Pine Site 4: Pine Savanna Site 4: Salt Marsh Site 5: Brackish Marsh Site 5: Control Axis 2 Site 5: Freshwater Marsh Site 5: Maritime Pine Site 5: Pine Savanna Site 5: Salt Marsh Axis 1 Stress value= 0.17, based on absence/presence data Effects of Site on Composition
• PERMANOVA: Pseudo-F = 31.05, p < 0.0010
% Sand % Silt % Clay Mean % Soil Moisture Salinity (ppt) Conductivity(µs/cm) Site 1 69.8 19.6 10.6 28.3 0.3 568.0 Site 2 63.8 23.6 12.6 40.0 0.1 199.2 Site 3 62.4 29.6 8.0 29.2 0.2 388.5 Site 4 66.4 25.0 8.6 33.5 0.1 221.1 Site 5 56.5 30.9 12.6 24.3 0.2 442.9 Effects of Propagule Sod on Composition
• PERMANOVA: Pseudo-F = 3.06, p < 0.0010
Maritime pine Control Salt marsh Brackish marsh Fresh marsh island
Salt marsh 0.001
Brackish marsh 0.104 0.081 Fresh marsh 0.001 0.001 0.001
Maritime pine island 0.001 0.001 0.001 0.105
Pine savanna 0.008 0.008 0.007 0.001 0.001 Recruitment of Target Species
• Targets emerge both from seed and resprout, most are indicator species from seed bank assessment
Vegetation Zone Total # of Target Most Abundant Species % of Subplots (Across Species all 5 sites) Salt Marsh 2 Spartina alterniflora 15 Distichlis spicata 5 Brackish Marsh 1 Juncus roemerianus 40 Freshwater Marsh 5 Sabatia stellaris 45 Panicum virgatum 15 Maritime Pine Island 6 Spartina patens 85 Pinus elliottii 20 Scirpus lineatus 15 Wet Pine Flatwood 17 Aristida beyrichiana 45 Andropogon glomeratus 35 Lachnanthes caroliniana 30 Aletris lutea 20 Lycopodiella prostrata 15 Sarracenia alata 5 Distribution of Desirable Taxa
• Mean number of taxa in each functional category varies significantly across sod treatment types and sites • More target species in freshwater and wet pine flatwood plots than in control and salt marsh plots
20 a a Generalist Species a Alien Species a a ab Target Species a a 15 c bc a Noxious Species c a ab ab ab a b 10
Mean # of taxa/plot # of Mean 5
a a a a a a 0 C S B F MP PF Sod treatment type Summary of Results
• Species diversity and richness increased, and noxious species greatly reduced on all treated areas
• Propagule sods resulted in “hybrid” communities (Hobbs et al. 2009), containing generalist and alien species, as well as indicator species from historic communities
• A viable propagule source for restorations, containing taxa with diverse life histories and environmental tolerances (Pywell et al. 1995, Brown and Bedford 1997, Anderson and Cowell 2004)
• Freshwater marsh and maritime pine island sods best suited to the restoration sites – seaward species can tolerate inland conditions
• Local environmental conditions and proximity to ruderal source populations also drive community composition/dynamics
Is anticipatory restoration feasible?
• Yes, but… – Differing degrees depending on species and background environmental change – Short-term success of propagule bank application in instilling diverse species into degraded ecosystems, long-term storage important
• Rate and direction of underlying abiotic change are key drivers – Communities that are sequentially arranged along strong environmental gradients may be easiest – Lack of spatial contiguity may require assisted colonization – Exotic species
• Use of common futuristic garden experiments at natural ecotones in the landscape to ease transition for vulnerable communities and enhance landscape fluidity (Manning et al. 2009) as climate change and sea-level rise alter site conditions Acknowledgements
• Brooks Davey • Jeni Miller • Scott Phipps • Scott Schuette • Joey Weber
Gulf of Mexico Questions?
Weeks Bay NERR
Gulf of Mexico Vegetation-Seedbank Comparison
• Procrustes RMS Residual (mean distance between parent vegetation and SB, goodness of fit)
1.5 Procrustes RMS Residual =0.5559 V SM V LB 1.0 V UB V FM V MPI 0.5 V WPF S SM
Axis 2 Axis S LB 0.0 S UB S FM S MPI -0.5 S WPF
-1.0 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 Axis 1