Life on the edge: Abiotic stress and competitive exclusion affect plant communities across a subalpine edaphic gradient

Ziyi Wang1, Willem Schep2, Hanan Abels-Sullivan3, Casey Kwok4

1University of California, San Diego; 2 University of California, Merced; 3 University of California, Los Angeles; 4 University of California, Berkeley

ABSTRACT

Abiotic stress and heavy each limit the assemblages of plant communities, defining their composition as well as their boundaries. One plant governed by stress and another by competition may abut one another, forming a biotic gradient called an . In California’s White Mountains, this gradient occurs across the meeting point between a calcareous (dolomite) and non-calcareous (quartzite) soil types. In this study, we tested the effects of this gradient on plant community compositions and diversity, surveying plant communities across an edaphic gradient from quartzitic to dolomitic substrates. We found that the higher presence of sagebrush within quartzite decreased diversity while increased rockiness in dolomite decreased plant cover. Diversity was highest in the ecotone between the two , potentially due to heavy competition from sagebrush on one side and the increase in edaphically stressful dolomite rock on the other. In addition, the ecotone plant community was unique from those on either soil type. These findings provide insight into on subalpine plant communities, and the nuances by which edaphic conditions produce these effects.

Keywords: ecotone, subalpine, California, edge effects, dolomite

INTRODUCTION have favorable habitats for plant growth, where any limitations on an individual's Plant communities can vary widely in growth is a result of other plants competing diversity. The diversity of these communities for the same resources. Competitive abilities is understood to be limited by three general between plants are never completely equal, factors: stress, competition, and and often a competitive dominant emerges . In a particular geographic area, which comes to exclude other species from one of these factors act as a dominating the landscape, such as mussels in the force limiting the number of colonists, California intertidal coast, which form a structuring that plant community. monoculture under removal of Pisaster Communities structured by competition will predators (Paine 1973). In areas where

CEC Research | https://doi.org/10.21973/N3M65V Summer 2018 1/8

conditions are not as favorable, stress on which only native grasses can tolerate. plants imposed by the physical environment When nutrients are added to these shapes communities, such as saline marshes serpentine soils, extremely competitive non- in estuaries. Plants taken in the most saline native grasses are able to move in and areas of the marsh thrive when placed in dominate the serpentine community freshwater, but only in the absence of (Huenneke et al. 1990). Essentially, regions neighbors, indicating stress from salt of relatively high environmental stress are concentration rather than competition is more resilient against invasion by prevalent in such an (Crain et al. competitive species. 2004). In these places, plants trade off In the subalpine White Mountains of competitiveness for tolerance against California, a soil gradient exists between environmental factors. patches of quartzite and dolomitic limestone are not isolated islands, derived soils, both of which greatly dictate however, and one plant community plant community. On dolomitic soil patches, governed by stress and another by plant cover is sparser and an open forest of competition may abut one another, forming bristlecone pines (Pinus longaeva) is a gradient from one limitation to the other widespread. Rocky outcrops are common as (Crain et al. 2004). Where one ecosystem a result of little soil development. Gradients blends into another, both biotically and between soils can range from one to several abiotically, this transition zone is referred to meters wide. On quartzitic soil, various as an ecotone. Because an ecotone is sagebrush (Artemisia spp.) acts as a influenced by both of the abutting competitive dominant, forming denser ecosystems, it will often have unique subshrub cover (Lloyd and Mitchell, 1973). characteristics. This is known as the edge Dolomite hosts poor edaphic, or soil-related, effect, first coined by Leopold in 1933. One conditions which are much lower in plant- of the most universal edge effects is a higher required nutrients and higher in calcium from overlapping carbonate than quartzitic soils, engendering communities (Harris, 1988). That said, a stress for the community adapted to ecosystem edges may also have negative inhabit them (Neely and Barkworth, 1984). effects on specific populations, especially Conversely, the sagebrush hosted by the when the edges are not of natural origin relatively macronutrient-rich quartzitic soil (Gates and Gysel, 1978). Because physical are tenacious competitors and can force and biotic factors may be different from the some less competitive species to grow on surrounding communities, are steep slopes and outcrops (Marchand 1973). often distinct from either ecosystem as well Between these two stressors exists the (Ries et al. 2004). This makes ecotones ecotone. unique areas of . The study observed how the plant One example of an ecotone across which a community and diversity changed across stress-competition gradient exists is in gradients of dolomite to quartzite soils California grasslands that enter serpentine- where patches abutted one another. We derived soils. Serpentine rock is low in hypothesized that, like the serpentine soils bioavailable nutrients, acting as a stress of the California grasslands, dolomite and

CEC Research | https://doi.org/10.21973/N3M65V Summer 2018 2/8

quartzite would form a stress to competition of approximately 10 km. The other 5 sites gradient in which dolomite's edaphic were on 3 smaller dolomite outcrops. In conditions would produce stress limitations, terms of substrate, 14 sites were quartzite, 2 while sagebrush hosted by quartzite would were a mixture of quartzite and granite, and impose competitive limitations. We also 1 was granite. Sites containing granitic soils hypothesized that the ecotone formed by were treated the same as quartzite soils as this physical gradient would allow for both of these non-calcareous soils have been increased diversity by acting as a refuge from found to foster the similar plant both of these limitations, hosting higher communities (Marchand 1973). To plant diversity. Finally, we hypothesized that standardize our sites, we chose 8 south because of a change in both stress and facing slopes, (aspect: 150°–210°), and 9 competition within a narrow area, the plant north facing slopes, (aspect: 330°–30°). All community present within the ecotone sites were at least 150 m away from each would be unique from either surrounding other and much bigger than 35 m x 10 m in community. size. All data was collected between August 2–5, 2018. METHODS 2.3 Study Design 2.1 Natural History of the Study System In order to account for abiotic variables, The study was conducted at the subalpine we measured elevation, slope, aspect and zone of the White Mountains in eastern ecotone width at each site. Then, to California, the northernmost part of the investigate the change in plant community White-Inyo Range. Elevation ranged from across the substrate gradient, based on the 3000 to 3500 meters. The landscape is its visible edges produced by change in rock mainly composed of sagebrush scrub atop types, we identified three zones at each site: quartzite and granite-based soils interrupted the dolomite, transition and quartzite zone. with dolomite outcrops and patches hosting A main vertical transect was placed parallel bristlecone pine forests. The dolomite to the slope, across the transition zone and usually occurs on top of the quartzite and can 15 m into both the dolomite zone and the be clearly distinguished by their white color. quartzite zone. We then placed 5 parallel transects perpendicular to the vertical 2.2 Site Selection transects. One at both ends of the vertical transect, one in the middle of the transition Our study focused on the plant zone, and one at both edges of the transition communities across the transition zones zone. To record richness, we used a 1 m x 10 between sagebrush dominated quartzite soil m belt transect on the horizontal transects, and predominated by dolomite patches. We species within the belt transect was counted haphazardly sampled selected 17 sites, and identified. To record evenness, plant where we observed a distinct transition cover of each species was measured along all between quartzite and dolomite. 12 of our horizontal transects using the point sites were surrounding a main patch of intercept method, where sample was taken dolomite patch spamming a distance

CEC Research | https://doi.org/10.21973/N3M65V Summer 2018 3/8

every 0.5 m for 10 m along the transect used as the metric for plant diversity. As we (Figure 1). When there was no plant present expected the diversity to be highest in the at the specific point, the substrate is transition zone than its surrounding recorded as either rock or soil. From our data quartzite and dolomite zone, quadratic on richness and evenness, we then regression was performed to examine the calculated Shannon diversity index (H’) for plant diversity across the five different all five horizontal transects of each site. locations along the main transect. ANOVA test was also used to explore the variation of sagebrush and rock cover across those five locations. In addition, linear regression was performed to determine if there was a correlation between plant diversity and abiotic (elevation, slope, aspect, transition zone width, soil/rock ratio) and abiotic factors (total sagebrush cover) at all 17 sites.

RESULTS

We found no effect of elevation, slope, aspect, and transition zone width on the plant diversity (Table 1). Among the five locations, quartzite, transition, and dolomite zone differed greatly from each other in plant composition, but the transition zone did not lay in a gradient between the quartzite and dolomite zones (Figure 2). In addition, the transition zone had the highest plant diversity compared to both quartzite 2 Figure 1. Transect methods. Five horizontal 10 m and dolomite zones (N = 17, R = 0.64, P = transects were placed perpendicular to the vertical 0.02; Figure 3). However, we did not find any transect, which extends 15 m into the Dolomite specific plant species that only appeared in substrate on one end and 15 m into Quartzite the transition zones. substrate on the other, passing through the transition Total sagebrush cover decreased across zone. quartzite to dolomite zones (Figure 4). In the 2.4 Statistical analyses quartzite sites, as sagebrush cover increased, Shannon Diversity Index (H’) JMP 14 software was used for all statistical decreased (N = 17, R2 = 0.21, P < 0.001; analyses. In order to determine the Figure 5). On the other hand, rock cover similarities and differences in plant increased across quartzite to dolomite zones communities between locations, we (Figure 6). In dolomite sites, total performed multivariable discriminant on all plant cover, both including and excluding 17 sites. Shannon diversity index (H’) was

CEC Research | https://doi.org/10.21973/N3M65V Summer 2018 4/8

Table 1. Abiotic variable effects on plant diversity. All site-specific abiotic variables, including elevation, slope, aspect and transition zone width showed no correlation with plant diversity.

Figure 3. Diversity in relation to location. The y-axis represents the Shannon Diversity Index, (H’), while the x-axis represents location of the transect, with (1) representing Quartzite, (2) Quartzite Transition, (3) Transition, (4) Dolomite Transition and (5) Dolomite respectively.

Figure 2. Multivariable discriminant analysis. Points A through E represent transects taken along a transition gradient from primarily quartzite substrate (A) into dolomite substrate (E). Points labeled A (red dots) indicate transects 15 m within the quartzite substrate, points labeled B (green triangles) indicate transects at the edge of quartzite substrate and transition zone, points labeled C (blue squares) indicate the center of the transition zone, points labeled D (gold asterisks) represent the dolomite transition edge, and finally points labeled E (cyan pluses) indicate transects 15 m within the dolomite substrate. Circles are the normal ellipse region Figure 4. Total sagebrush cover in terms of location. estimated to contain 50% of the data points for each The y-axis represents total sagebrush cover in transect. Separation of circles A, B, and E indicates percentage. X-axis represents locations along the that the plant communities surveyed within these vertical transects, with (1) representing Quartzite, (2) transects are not only different from one another, but Quartzite Transition, (3) Transition, (4) Dolomite also do not change in a stepwise gradient across the Transition and (5) Dolomite respectively. Levels not transition zone. Rather, communities within this zone connected by same letter are significantly different. are distinct from those on either side. Bars represent + 1 S.E.M.

CEC Research | https://doi.org/10.21973/N3M65V Summer 2018 5/8

Figure 5. Rock cover correlation with plant cover in Figure 7. Rock cover correlation with plant cover in dolomite site. The y-axis represents plant cover, dolomite site. The y-axis represents plant cover, while the x-axis represents rock cover. The dotted line while the x-axis represents rock cover. The dotted line represents plant cover without sagebrush, while the represents plant cover without sagebrush, while the dashed line represents plant cover with sagebrush. As dashed line represents plant cover with sagebrush. As the rock cover increases, both the plant cover with the rock cover increases, both the plant cover with sagebrush and without sagebrush decreased. sagebrush and without sagebrush decreased. sagebrush, were negatively correlated withwith rock cover (with sagebrush: N = 17, R2 = 0.43, P < 0.01; without sagebrush: N = 17, R2 = 0.41, P < 0.01; Figure 7).

DISCUSSION

The transition zone located between sagebrush dominated quartzite and the rocky dolomite had the highest plant diversity as we predicted. However, the plant composition of the transition zone did not lay in between the quartzite and dolomite zones in multivariable discriminant Figure 6. Rock cover in terms of location. The y-axis analysis, meaning it fostered a distinct plant represents total rock cover in percentage. X-axis represent locations along the vertical transects, with community separated from those at either (1) representing to Quartzite, (2) Quartzite Transition, extreme. This pattern supported our (3) Transition, (4) Dolomite Transition and (5) hypothesis that quartzite to dolomite Dolomite respectively. Levels not connected by same ecotone provided available niches for a letter are significantly different. Bars represent + 1 more diverse plant community. We S.E.M. attributed the unique assemblage of the transition zone to the decrease in competition from sagebrush and lower physical stress from dolomite rock.

CEC Research | https://doi.org/10.21973/N3M65V Summer 2018 6/8

The negative correlation between (Jensen 1990). Furthermore, sampling in the sagebrush cover and plant diversity in the core of the dolomite and quartzite quartzite zone aligned with Marchand’s ecosystems, far from the edge, would allow study that when sagebrush was dominant it a better understanding of these acted as an exclusive competitor (Marchand communities’ compositions outside the 1973). In addition, the decrease in physical effects of the ecotone. In addition, future stress may also contributed to the favorable studies that would offer deeper analysis of conditions of the transition zone. As rock the effects of abiotic dolomite-based cover decreased, plant cover increased stressors, sagebrush competition, and their across the dolomite to the transition zone, relationships to edge effects could include implying increased . However, nutrient addition and sagebrush removal evidences for the abiotic limitation on the experiments, as well as a study of changes in plant community of dolomite were difficult substrate-based species morphology to to determine from our methods, but the learn how different organisms respond to trend of decreased species diversity within various environments and nutrients. predominantly dolomite substrate Ecotones pose a challenge to scientific suggested edaphic limitation. Dolomite, as a research due to their variability and calcareous substrate, tends to be more heterogeneity (Ries, Fletcher, Battin, Sisk alkaline and contain fewer nutrients than 2004). However, the concept of the ecotone quartzite (Neely and Barkworth 1984). With as a vital aspect of allows for a more the decreased competition from the comprehensive picture on community sagebrush and physical stress, the transition interaction. As the plant community of zone hosted a distinct plant community that ecotone behaves differently from its differed from its adjacent quartzite and surrounding ecosystems, communities need dolomite zones. to be studied as heterogeneous entities, Furthermore, our results supported the rather than singular units. In addition, the findings of M.E. Jensen (1990) which found edge effects on plants are hard to study on a elevation and aspect had no impact on plant community level because individual plants diversity and were not factors determining will respond to specific biotic and abiotic sagebrush community type or composition. factors (Marchand 1973).A factor that Instead Jensen found that soil moisture and controls the distribution of one species is not temperature had a greater impact on changes necessarily correlated with those that affect in communities dominated by sagebrush. other species. These results add another To expand on this work, extending the perspective to our understanding of lesser sample period, expanding the range of studied ecosystem edge dynamics. The sample sites and increasing number of findings of higher diversity within the replicants would allow for greater ecotone boundary support similar findings explanatory power. Additional testing might by numerous studies (Ries, Fletcher, Battin, be included for factors such as rock Sisk 2004). Our study offered a single fractionation, soil pH, soil moisture, and glimpse into understanding the similarities surface temperature, as these variables are and differences between both biotic and likely to show an effect on plant composition abiotic factors on a community.

CEC Research | https://doi.org/10.21973/N3M65V Summer 2018 7/8

ACKNOWLEDGMENTS Californian serpentine grassland. Ecology 71:478– 491. Research was based out of the UC NRS Jensen, M.E. (1990). Interpretation of environmental White Mountains Research Station at gradients which influence sagebrush community Crooked Creek, doi:10.21973/N3KM2J. We distribution in northeastern Nevada. Journal of wish to give special thanks to Dylan Range Management 43:161–167. Neubauer for help with plant identification and professors Tim Miller and Krikor Leopold A. (1933). Game Management. New York: Charles Scribner’s Sons. Andonian for their guidance. Lloyd, R. M., and R. S. Mitchell (1973). A flora of the REFERENCES White Mountains, California and Nevada. Berkeley: University of California Press. Crain, C. M., Silliman, B. R., Bertness, S. L., & Bertness, M. D. (2004). Physical and biotic drivers of plant Marchand, D.E. (1973). Edaphic control of plant distribution across estuarine salinity gradients. distribution in the White Mountains, Eastern Ecology 85(9):2539–2549. California. Ecology 54:233–250.

Farina, A. (2009). The Ecotones. Ecology, Cognition Neely, E.E., Barkworth, M.E. (1984). Vegetation on and Landscape pp. 83–93. Soils derived from dolomite and quartzite in the Bear River Range, Utah: a comparative study. Gates, J. E., & Gysel, L. W. (1978). Avian nest Bulletin of the Torrey Botanical Club 111:179–192 dispersion and fledging success in field-forest ecotones. Ecology 59(5):871–883. Ries, L., Fletcher Jr., R.J., Battin, J., Sisk, T.D. (2004). Ecological responses to edges: Harris, L. D. (1988). Edge effects and conservation of Mechanisms, models, and variability explained. biotic diversity. Conservation Biology 2(4):330–332. Annual Review of Ecology, Evolution, and Systematics 35:491–522 Huenneke, L.F., Hamburg, S.P., Koide, R., Mooney, H.A., Vitousek, P.M. (1990). Effects of soil resources Schlesinger, W.H., Raikes, J.A., Hartley, A.E., Cross, on plant invasion and community structure in A.F. (1996). On the spatial pattern of soil nutrients in desert ecosystems. Ecology 77:364–374

CEC Research | https://doi.org/10.21973/N3M65V Summer 2018 8/8