30 PARK SCIENCE • VOLUME 26 • NUMBER 2 • FALL 2009 Science Features

NPS/JOY MARBURGER

Exploring the infl uence of genetic diversity on pitcher plant restoration in Indiana Dunes National Lakeshore By Jennifer M. Karberg, Joy Marburger, and Margaret R. Gale

Figure 1. The northern CARNIVOROUS PLANTS OBTAIN NUTRIENTS through overcollection. As habitat and populations pitcher plant traps prey needed for growth through the breakdown of of the northern pitcher plant become increasingly in its tubelike pitcher structure where a suite insects, microbes, and small amphibians. The most rare, state and federal agencies are showing greater of microorganisms di- widely distributed carnivorous plant in North interest in conserving habitat and restoring plant gests prey and extracts nutrients needed for America is the northern pitcher plant (Sarracenia populations. plant growth. purpurea L., fi g. 1), whose range stretches from northern Canada to the midwestern United States, Indiana Dunes National Lakeshore protects one of and along the eastern U.S. coast south to the Gulf the few remaining populations of northern pitcher of Mexico. This species lives primarily in isolated, plant in the state (NatureServe 2007) (fi g. 3). This low-nutrient sphagnum moss bog and poor fen population, located at the Indiana Dunes Pinhook wetlands (fi g. 2). Though individual populations Bog property, is isolated within an extensively are large, typically containing more than 100 plants, developed landscape along the southern rim of the species is in decline because of the loss of its Lake Michigan east of Gary, Indiana. Consequently, specialized wetland habitat. The wetlands that host the national lakeshore has experienced declining the northern pitcher plant are in a perilous position, populations of the northern pitcher plant. Scientifi - often drained for development, mined for Sphag- cally informed management to restore this species num for the horticultural trade, or degraded by in- is crucial to its survival in this and other fragmented puts of road salt and lawn and agricultural fertilizer ecosystems. Planning and implementing successful runoff . Additionally, carnivorous plant enthusiasts restoration of plant populations requires knowl- prize this species and threaten population survival edge about how the plant functions ecologically, SCIENCE FEATURES 31

how it reproduces, what environmental qualities it 2003; Allendorf and Luikart 2007). Using seed or requires, and how populations relate to each other plants from distant populations that are planted genetically. Reestablishment of the pollination ser- with locally adapted populations to restore new

NPS/JOY MARBURGER vices provided by bees in Pinhook Bog will sustain sites could lead to reduced fi tness in off spring from the reproductive potential and genetic resilience of crosses between the locally adapted and distant pitcher plants (Dixon 2009). populations in the restoration site (Huff ord and Mazer 2003). Careful consideration and under- Figure 2. Though wide- standing of the genetic relatedness of populations spread in North America, become important in restoration when considering the northern pitcher Importance of genetics plant faces threats from supplementing plant populations with outside seed loss and fragmentation in restoration planning sources. of its specialized habitat: low-nutrient bog and Sphagnum bog and fen wetlands south of Canada poor fen wetlands. Indi- are typically isolated, separating populations of Consider these two examples of how genetic un- ana Dunes populations are genetically interme- the northern pitcher plant both physically and by derstanding can aid ecosystem management: diate between the iso- distance. This can lead to interesting conservation lated island population of Isle Royale and the issues regarding the genetic relationships among • Imagine many populations of a plant species continuously interbreed- diff erent pitcher plant populations. One of the key where pollen and seeds are continually being ing mainland popula- questions ecologists have tried to answer is how exchanged among the populations. These pol- tions of Apostle Islands and Pictured Rocks. fragmentation or isolation in the landscape aff ects len and seeds represent gene fl ow among the genetic variation and gene fl ow of plant populations populations, and the more gene fl ow that occurs, of a particular species. Wright (1969) developed the more similar the populations become to the island model of migration among populations, each other genetically. Where some event has which is frequently applied to landscape islands reduced, destroyed, or removed a plant popula- or fragmented natural areas such as the habitat of tion from a site, managers can potentially restore the northern pitcher plant. He hypothesized that the population by collecting and reintroducing within populations of a species, two important genetic forces function between isolated or island populations: (1) gene fl ow originating from out- side populations that maintain genetic relatedness Voyageurs NP Grand Portage NM Isle Royale NP between populations either through pollen or seed Keweenaw NHP exchange, and (2) genetic drift occurring within Apostle Islands NL Pictured Rocks NL

populations that create more genetically diff erent Saint Croix NSR MINNESOTA subpopulations (Allendorf and Luikart 2007). The Sleeping Bear Dunes NL Mississippi NRR more geographically isolated a population becomes,

the less gene fl ow will occur between populations, WISCONSIN increasing genetic drift and potentially creating MICHIGAN genetically distinct populations (Kimura and Weiss IOWA Indiana Dunes NL 1964; Allendorf and Luikart 2007). Essentially, the OHIO amount of divergence between two populations ILLINOIS INDIANA is a balance between gene fl ow homogenizing the populations and genetic drift diff erentiating the populations. The more isolated the populations, the less balanced these processes are, thus leading to Figure 3. This study assessed the population genetics increased genetic divergence. of pitcher plants in four national park units of the Great Lakes region shown by the red dots: Isle Royale National Park and Indiana Dunes, Pictured Rocks, and Eff ects of genetic isolation within populations Apostle Islands national lakeshores. The main goal of the study was to determine if abundant popula- complicate habitat and plant species restoration tions of pitcher plants in some parks could be used planning. Isolated populations experiencing genetic to supplement decreasing populations at Indiana Dunes National Lakeshore without compromising lo- drift and increased inbreeding can become locally cal population genetics or reducing successful local adapted to a particular site (Huff ord and Mazer adaptations. 32 PARK SCIENCE • VOLUME 26 • NUMBER 2 • FALL 2009

The amount seeds from any of the surrounding populations Genetic analysis of the of divergence because all of the populations are genetically western Great Lakes similar. Consideration of restoration sites may between two thus focus on those with similar ecological con- populations populations ditions, rather than the genetic component. We collected plant leaves for molecular genetic analysis to determine the genetic relatedness of is a balance • In the second scenario, many populations of a populations at each of the four sampled national between gene fl ow plant species are still spread across the land- parks: Apostle Islands, Indiana Dunes, Isle Royale, scape, but one population is isolated on an island and Pictured Rocks. The analysis indicated a strong homogenizing in a lake. This population is so isolated by dis- degree of clonal growth, created by asexual repro- tance from other populations that very little seed duction, evidenced by moderate within-population the populations or pollen exchange occurs between the island genetic diversity (Nei’s gene diversity = 0.3770  and genetic drift and the mainland; therefore, gene fl ow is very 0.1289). Clonal species tend to have lower overall low for this population. As gene fl ow is reduced, genetic diversity within a population because of their diff erentiating the populations can begin to become genetically nonsexual reproductive methods. The northern populations. diff erent from each other, and this is an illustra- pitcher plant has often been hypothesized as being tion of genetic drift. If the island population clonal (Schwaegerle and Schaal 1979). This moderate requires reintroduction, managers need to care- amount of genetic diversity was examined using an fully consider where to select seeds to restore analysis of molecular variance (AMOVA) to see how this population. Introducing seed from outside much variation exists within individual populations sources could lead to the loss of the local plants’ and how much exists between or among populations genetic uniqueness, and potentially decrease the at diff erent locations. The total variation explained success of restoration attempts. by genetic diff erences between populations located

in each park (FST = 0.2688, p < 0.0001) indicates that Fortunately, populations of northern pitcher plant 26.88% of the observed genetic diversity was ac- are still fairly abundant in other national park units counted for by population variation. In essence, our within the western Great Lakes region. We wished to sampled populations are 26.88% genetically diff erent assess whether some of these populations could serve from each other. To determine what this means in as potential seed sources for restoration of this plant terms of how the individual populations are related species within Indiana Dunes National Lakeshore. to each other, a visual analysis of the genetic dissimi- The northern pitcher plant has continuously exhib- larity of populations is necessary. Genetic distance or ited low genetic diversity within and among popula- genetic dissimilarity of populations can be graphi- tions throughout the eastern United States (Godt cally represented with a dendrogram or diagram tree and Hamrick 1996; Schnell 2002; Karberg and Gale that illustrates how closely related populations are 2006). This means that throughout its range the plant genetically. An examination of the visual patterns has consistently displayed low genetic diff erences be- of genetic distance between populations revealed tween individuals and between populations regardless genetic diff erences that distinguished Indiana Dunes of distance between populations. This would indicate and Isle Royale populations from the other two parks that most populations are genetically similar and seed (fi g. 4). This dendrogram graphically represents sources for restoration could be plentiful. genetic distance between individual populations and illustrates that Isle Royale populations, segregated on We wished to examine diff erences in genetic varia- a separate branch of the dendrogram, have the great- tion between the Pinhook Bog population located est genetic distance from the other park populations. within Indiana Dunes National Lakeshore and Apostle Islands and Pictured Rocks populations are three other national park units in the western Great closely intermixed and not well separated by genetic Lakes area—Isle Royale National Park (Michigan), distance. Indiana Dunes populations are genetically Apostle Islands National Lakeshore (Wisconsin), and intermediate between the isolated island population Pictured Rocks National Lakeshore (Michigan)—to of Isle Royale and the continuously interbreed- determine the possibility of using seed from regional ing mainland populations of Apostle Islands and sources for restoration at Indiana Dunes (see fi g. 3). Pictured Rocks. SCIENCE FEATURES 33

Natural and landscape and increases population inbreeding and genetic island populations drift, making the population more genetically dis- tinct from mainland populations. The two populations of the northern pitcher plant that exhibited the most genetic diff erentiation In evaluating the results of our genetic analysis, from the other populations are both fairly isolated which diff erentiated the landscape island population populations: the natural island population of Isle of Indiana Dunes from other intact populations, we Royale National Park is isolated within Lake Supe- think that larger expanses of fragmented landscape rior, and the landscape island population of Indiana may also provide a suffi cient barrier, reducing gene Dunes National Lakeshore is isolated within a fl ow to the Indiana Dunes population. The two highly human-altered, fragmented landscape. The nearest populations of the northern pitcher plant to genetic diff erentiation of these two populations as Indiana Dunes are located 45 miles (73 km) and 96 compared to mainland populations found in a non- miles (154 km) away from the study site (NatureServe fragmented landscape can potentially be explained 2007). Hypothetically, 45 miles (73 km) of frag- by habitat isolation. The natural island population mented landscape may represent the outer bounds ( Isle Royale) is located on an isolated island within of a fragmentation threshold for this plant, beyond Lake Superior that has lacked a physical connection which genetic diff erentiation among populations to the mainland since the last glaciation (Dorr and increases as gene fl ow proportionally decreases. Eschman 1970; Huber 1983). Northern pitcher plant seeds are not believed to be consumed by birds or The mechanism for genetic diff erentiation of the mammals, which would facilitate distribution over landscape island population may be related to gene water, but rather typically disperse close to the fl ow and distance, with Indiana Dunes receiving parent plant (Ellison and Parker 2002). The barrier little gene fl ow from outside populations because of of large stretches of open water may likely be the distance, thus increasing the action of genetic drift. reason for genetic diff erentiation of the northern Pollinators of the northern pitcher plant include pitcher plant populations on Isle Royale since the two bee species, Bombus affi nis and Augochlorella water barrier reduces the possibility of gene fl ow aurata, and one fl y species, Flecherimyia fl etcheri (Ne’eman et al. 2006). Observations suggest that specifi c pollinators are limited at the Indiana Dunes Pictured Rocks (1) site (A. Molumby, University of Illinois–Chicago, personal observation). Research has shown that Apostle Islands (1) wild bee pollinations are limited by habitat isola- tion; increased distance from suitable habitat leads to decreased bee visitations (Steff an-Dewenter Pictured Rocks (2) and Tscharntke 1999). This factor of limited pollen dispersal contributes to the landscape isolation of populations within both of the sampled “island” Apostle Islands (2) populations.

Indiana Dunes Indiana Dunes northern pitcher plant populations show a degree of genetic isolation from other na- tional park populations. This is a plant species that Isle Royale has historically shown a low degree of genetic varia- tion, so even small amounts of genetic diff erentia- Genetic Distance 0.00 0.37 tion could be signifi cant. Understanding how popu- lations with this degree of isolation would respond Figure 4. This dendrogram illustrates the genetic distance between the six sampled to restoration through introduction of outside seed populations of northern pitcher plant. The Isle Royale and Indiana Dunes branches sources or plants is not yet fully understood. The are separated from the other two parks, indicating unique genetic patterns in these populations. The more distant or separate the branch, the more genetically accepted general standard among restoration ecolo- distant the populations. gists is to utilize seeds or vegetative plant material 34 PARK SCIENCE • VOLUME 26 • NUMBER 2 • FALL 2009

Science Features

collected within a 50-square-mile (12,950 hectares) of outside seed sources for restoration of declined area immediately surrounding the restoration site. northern pitcher plant populations? This study However, this generalization is anecdotal and needs alone is not enough to conclusively predict the

NPS/JOY MARBURGER to be examined scientifi cally, given what we have results of introducing outside genetic sources but learned about clonality and genetic diff erentia- the diff erences are large enough to argue for cau- tion due to geographical isolation in the northern tion and further exploration before using outside pitcher plant. To examine restoration possibilities sources. The northern pitcher plant has consistently Figure 5. Investigators at Indiana Dunes, a common garden experiment shown smaller amounts of genetic diversity across are growing pitcher could be undertaken to examine responses of plants its range. It’s a clonal plant, which typically reduces plants from seed in greenhouse conditions and seeds from all four national parks planted side genetic diversity within a species, and yet signifi cant at Indiana Dunes for by side in a single site (fi g. 5). Examining the growth diff erences were still detected between Indiana wetland restoration. These seedlings take 3 responses and interbreeding fi tness of plants from Dunes populations and populations in other parks. to 5 years to reach ma- these diff erent populations on one substrate type Before outside introduction, a common garden turity, which may not be desirable for restoration would help determine whether observed genetic study should be conducted to determine the infl u- plans. diff erences are ecologically signifi cant. ence of local genetics versus local environment on northern pitcher plant ecology.

Lessons for restoration In the meantime, what are the options to re- store pitcher plant populations in Indiana Dunes The northern pitcher plant has historically exhib- Na tional Lakeshore? One is to improve habitat ited low genetic variability, with populations very conditions for pitcher plant establishment. Man- genetically similar to each other often across great agers have begun removing blueberry plants to distances. However, this study did document ap- create more open habitat suitable to the northern preciable genetic diff erentiation between isolated, pitcher plant. To hasten restoration at Pinhook Bog, island populations and mainland populations. managers can attempt to increase seed set through Currently, northern pitcher plant populations in hand pollination, promoting local seed produc- Indiana Dunes show some genetic distinction from tion. Transplanting vegetative clones can also be other mainland populations. Genetic isolation conducted within the bog. Introduction of associ- caused by artifi cial geographic barriers, such as ated bee pollinators and their required habitat is urban and agricultural development, limits cross- also recommended. Indiana Dunes pitcher plant pollination and thus possibly reduces genetic varia- populations do show some degree of genetic isola- tion in the Pinhook Bog population. A decrease in tion from the other national park populations, and pollinator populations may also be contributing to the consequences of using outside plants and seeds limited gene fl ow. Previous researchers have noted to supplement restoration eff orts are questionable. that the habitat for bumblebees, the natural pol- More research, especially using common garden linators of the pitcher plant, is lacking in Pinhook experiments, is needed to determine if introduced Bog (A. Molumby, University of Illinois–Chicago, plants from other regions of the Great Lakes can personal communication). Another factor contrib- successfully reproduce without reducing the genetic uting to decline of the pitcher plant populations is integrity of northern pitcher plant populations in the expansion of highbush blueberries (Vaccinium Indiana Dunes. The results from plants grown from corymbosum) in the bog. The pitcher plant is shade- clones would be faster than from seed-derived intolerant and thus plant growth is limited because plants, which would not be available for 3 to 5 years, of shading eff ects of blueberry plants. (Mystery given the slow maturation from seed establishment surrounds the origin of blueberry plants in Pinhook to adult plants of this species (fi g. 5). In the mean- Bog. Anecdotal evidence suggests that a previous time, increasing habitat quality and facilitating seed property owner planted them for agriculture prior set of native populations may be the best way to to authorization of the national lakeshore in 1966.) restore pitcher plant populations today at Indiana The major question is: Is this amount of genetic dif- Dunes. ferentiation or genetic identity observed in the Indi- ana Dunes populations enough to preclude the use SCIENCE FEATURES 35

Wright, S. 1969. Evolution and the genetics of populations. References Volume 2. University of Chicago Press, Chicago, Illinois, Allendorf, F. W., and G. Luikart. 2007. Conserving USA. global biodiversity? Conservation and the genetics of populations. Blackwell Publishing, Oxford, UK. For further information Dixon, K. W. 2009. Pollination and restoration. Science • http://www.mwpubco.com/PitcherPlants.htm 325:571–572. • http://www.biggerbooks.com/bk_detail. spx?isbn=9780881927924 Dorr, J. A., Jr., and D. F. Eschman. 1970. Geology of Michigan. The University of Michigan Press, Ann Arbor, • http://www.sarracenia.com/faq.html Michigan, USA. http://www.honda-e.com/ • http://en.wikipedia.org/wiki/Sarracenia_ Ellison, A. M., and J. N. Parker. 2002. Seed dispersal purpurea#Taxonomy and seedling establishment of Sarracenia purpurea (Sarraceniaceae). American Journal of Botany 89:1024– Funding 1026. This work was funded primarily through research Godt, M. J. W., and J. L. Hamrick 1996. Genetic structure grants provided by the Great Lakes Research and Ed- of two endangered pitcher plants, Sarracenia jonesii ucation Center, Great Lakes Northern Forest Coop- and Sarracenia oreophila (Sarraceniaceae). American erative Ecosystem Studies Unit, and the National Park Journal of Botany 83:1016–1023. Service Cooperative Conservation Initiative. Michigan Huber, D. J. 1983. The role of cell wall hydrolases in fruit Technological University and the DeVleig Foundation softening [Texture changes, food quality.] Horticultural provided technical and researcher support. Reviews 5:168–219. Hufford, K., and S. J. Mazer. 2003. Plant ecotypes: Genetic differentiation in the age of ecological restoration. About the authors

Trends in Ecology and Evolution 18:147–155. MARBURGER NPS/JOY Jennifer M. Karberg1 Karberg, J. M., and M. R. Gale. 2006. Genetic diversity and is the research supervisor distribution of Sarracenia purpurea (Sarraceniaceae) in the western Lake Superior basin. Canadian Journal of for the Science and Botany 84:235–242. Stewardship Department of the Nantucket Kimura, M., and G. Weiss. 1964. The stepping stone model Conservation Foundation, of population structure and the decrease of genetic Nantucket, Massachusetts. correlation with distance. Genetics 49:561–576. This work was completed NatureServe. 2007. NatureServe explorer: An online as a portion of her PhD encyclopedia of life. Version 7.1. Updated 9 February research with Michigan Technological University, 2009. Available at http://www.natureserve.org/explorer. School of Forest Resources and Environmental Ne’eman, G., R. Ne’eman, and A. M. Ellison. 2006. Limits Science, Houghton, Michigan. to reproductive success of Sarracenia purpurea (Sarraceniaceae). American Journal of Botany 93:1660– Joy Marburger is the research coordinator with the 1666. Great Lakes Research and Education Center, Indiana Schnell, D. E. 2002. Carnivorous plants of the United States Dunes National Lakeshore, Porter Indiana. and Canada. 2nd edition. Timber Press, Portland, Oregon, USA. Margaret R. Gale is dean of the School of Forest Resources and Environmental Science at Michigan Schwaegerle, K. E., and B. A. Schaal. 1979. Genetic Technological University, Houghton, Michigan. variability and founder effects in the pitcher plant Sarracenia purpurea L. Evolution 33:1210–1218.

Steffan-Dewenter, I., and T. Tscharntke. 1999. Effects of 1 Corresponding author Jennifer M. Karberg habitat isolation on pollinator communities and seed Nantucket Conservation Foundation set. Oecologia 121:432–440. 118 Cliff Rd Nantucket, MA 02554