Resisting Invasion on Tussocks

Revisiting invasion on tussocks: Relating environmental variables to invasion

Martin Genova1, Conall Gaffney2, Anson Pang3

1University of California, Santa Cruz, 2University of California, Berkeley 3University of California, Los Angeles

ABSTRACT

Classic invasion theory predicts that higher species diversity increases the resistance of a community to invasion by non-native species. However, studies that challenge this theory propose that environmental variables that covary with native diversity play an important role in determining invasibility in communities. One way that environmental variables can act to determine community invasibility is by facilitating propagule pressure of non-native species. Here, we investigate how environmental variables, through the mechanism of propagule pressure, directly act to influence the invasion success of non-native plant species on riparian tussock communities in the South Fork Eel River (Mendocino County, CA). We found that distance to upstream tussocks influenced the number of non-native species on tussocks, independent of native species richness. The results of this study support the idea that environmental variables interplay with native diversity to determine community invasibility.

Keywords: invasion, tussocks, propagule pressure, biotic resistance, diversity/invasibility hypothesis

INTRODUCTION invasion need to be investigated in order to develop optimal management strategies As humans have spread species beyond across different systems. their native range, the impact of their One influential theory that has predicted introductions have negative economic and the invasibility of plant communities is the ecological implications (Pejchar & Mooney diversity/invasibility hypothesis, first 2009). Understanding the factors that developed by Elton (1958). The hypothesis influence the establishment of invasive posits that more diverse plant communities species in natural communities can help are less likely to be invaded and established predict the extent of invasion in natural invaders will have reduced success. This systems (Fridley et al. 2007) This theory has been supported by theoretical understanding can better inform and mathematical models (Case 1990, management strategies that maintain MacArthur 1970), and microcosm field biological diversity and ecosystem functions experiments where native species diversity (Fridley et al. 2007, Levine et al. 2003). is experimentally manipulated (Knops et al Therefore, mechanisms that influence

CEC Research | https://doi.org/10.21973/N3WM1Z Spring 2019 1/9

1999, Levine 2000, Symstad 2000). For strength of the diversity/invasibility example, the performance of an invasive hypothesis and its effect on shaping plant annual weed, narrowleaf hawksbeard communities (Levine 2000, Von Holle 2005). (Crepis tectorum), was found to be One mechanism influenced by negatively correlated with native species environmental variables that can determine richness in experimental Minnesota the success of invaders is propagule grassland plots (Naeem et al. 2000). These pressure (Colautti et al. 2006, Simberloff findings implicate that maintaining biodiversity 2009, Von Holle and Simberloff 2005). In in natural communities is necessary to the context of plants, propagules can be reduce invader establishment and success. vegetative cones or seeds that can be While the diversity/invasibility hypothesis carried by biotic (e.g. herbivores) or abiotic has been well supported, the pertinence of (e.g. wind, water) vectors. Propagule the hypothesis to what actually occurs in pressure is a measurement that incorporates plant communities is debated among propagule size and the temporal and spatial ecologists (Tilman 1999, Enserink 1999). patterns of propagule arrival to predict the There have been several observational likelihood that a species establishes in a studies that have refuted the applicability of new community (Simberloff 2009). The the diversity/invasibility hypothesis to Theory of Island Biogeography (Macarthur natural communities, often finding invader & Wilson 1967) correlates propagule success to be positively correlated with pressure with environmental variables such native species richness (Keeley et al. 2003, as the size of an island and the distance to Planty-Tabacchi et al. 1995, Wiser et al. mainland to determine the rate at which 1998). However, critics of these species establish. The same general idea observational studies point out that the can be applied to plant invaders, in which mechanisms driving non-native success are environmental variables can influence the in reality environmental variables that likelihood that propagules will arrive in new covary with native species richness, making habitat patches (Simberloff 2009). it appear that diversity increases non-native However, the way that propagule pressure invasion success (Levine & D’Antonio 1999, acts to facilitate invasion success can vary Stohlgren 1999). Additionally, there has among ecosystems (e.g. Von Holle & been a growing body of experimental Simberloff 2005). Therefore, investigating evidence that environmental variables the environmental variables that can act to determine invader success more than increase propagule pressure in a given plant native species richness (Luo et al. 2018, system can help ecologists predict the Towers & Dwyer 2018, Von Holle 2005). For success of invasion. example, the abundance of a non-native In this study, our aim was to elucidate the invasive, tropical whiteweed (Ageratum environmental variables that affect the conyzoides), was found to be more strongly invasibility of tussocks on the South Fork Eel correlated with the organic matter and River in Northern California. Tussocks are nitrogen content in the soil than with native numerous and discrete units of habitat that species cover (Luo et al. 2018). With this in allow us to investigate how environmental mind, taking environmental variables into variables act to facilitate the establishment account is necessary to determine the of native and non-native species. Levine

CEC Research | https://doi.org/10.21973/N3WM1Z Spring 2019 2/9

(2000) found that the number of non-native METHODS species on tussocks along a 7 km stretch of the South Fork Eel River increased with 2.1 Study System native species diversity at the community level (along the river). However, when We conducted our research on South Fork native species richness was manipulated at Eel River in Angelo Coast Range Reserve the neighborhood level (within tussocks), (Mendocino County, California) May 7–11, he found that the opposite was true. Levine 2019. The focus of our research was on (2000) suggested that the positive tussocks, which are small, grassy islands relationship between non-native and native formed in the river by torrent sedge (Carex species richness on tussocks at the nudata) (Levine 2000). Singular C. nudata community level could be due to physical are perennial plants that grow in clumps factors that covary with native species richness. and catch organic matter in the river, The objective of our study was to creating habitat for other plants. We investigate environmental variables that surveyed tussocks in 14 groups (Figure 1) in drive the observed positive native/non- which groups were defined as clumps of at native species richness relationship at the least nine individual tussocks. All tussocks community level. The results of this study within a group were within 5 m from can shed light on the importance of native another tussock within the group. diversity on invader success on tussocks. 2.2 Field Survey Using propagule pressure as a mechanistic framework driving invader establishment on Within each tussock group, we randomly tussocks, we hypothesized that (1) tussocks selected nine tussocks to measure several closer to the front of groups of tussocks environmental variables that could drive that (2) had a larger area, (3) were propagule pressure on tussocks. To see how experiencing a lower stream velocity, (4) propagules from upstream influence were closer to upstream tussocks, and (5) tussock native and non-native species were closer to the stream shore would have richness, we looked at (1) the location along a greater number of both native and non- group of tussocks, (2) tussock area, and (3) native species richness. We believe these the stream velocity for each tussock. patterns would occur because such tussocks Propagules from other tussocks was are likely to receive more seeds carried investigated by looking at (4) the distance along the stream or arriving from shore, to upstream tussocks. The source of and thus experience a higher propagule propagules from the shore was examined as pressure. In this study, we also wanted to the (5) distance to stream shore. The answer whether environmental variables location along a tussock group was directly influence invasion success on calculated as a percentage that each tussocks independent of native species tussock was along the total length of the richness. The answer to this question would group (Figure 2). For distance to upstream allow us to disentangle the environmental tussocks, we measured the distance a variables that could covary with native tussock was from the two closest upstream diversity yet directly influence the invasibility tussocks that were within 45 degrees from of tussocks.

CEC Research | https://doi.org/10.21973/N3WM1Z Spring 2019 3/9

a vertical line drawn directly upriver (Figure obstruction 3 m upstream from a tussock, 2). Stream velocity was measured by placing shorter distances of 1.5 m or 1 m were a standard-sized ping pong ball in water 3 m used. The area of each tussock was directly upstream from a tussock and timing calculated as the area of an ellipse (� x how long the ping pong ball takes to travel radius 1 x radius 2), where radius 1 was over the 3 m. We measured the velocity parallel to shore and radius 2 was three times and took the average of the perpendicular to shore. measurements. When there was an

Figure 2. Model diagram showing how tussock location along group and distance to upstream tussocks were measured. Green ovals represent tussocks in the South Fork Eel River in Angelo Coast Range Reserve (Mendocino County, CA). “a” represents the total distance of a group of tussocks. “b” represents the distance of an individual tussock from the tussock at the front of a group of tussocks. Figure 1. Map of study sites along the South Fork Location along the tussock group was calculated as a Eel river in Angelo Coast Range Reserve percentage that each tussock was along the length (Mendocino, CA). The white star represents the of the group (b divided by a). “c” and “d” represent location of Angelo Coast Range Reserve. Black circles the distance to the two closest upstream tussocks represent groups of tussocks observed in our study. from an individual tussock. “e” represents a vertical line drawn directly upriver from an individual tussock. Distance to upstream tussocks was defined as the average distance to the two closest upstream tussocks (c and d) that were within 45 degrees from a vertical line drawn directly upriver (e).

CEC Research | https://doi.org/10.21973/N3WM1Z Spring 2019 4/9

For each selected tussock, we noted the 0.0001; Table 1a). Distance to upstream presence of all plant species. We tussocks, stream velocity, and distance distinguished which plants were native along group were all negatively correlated species and which ones were invasive. From to total plant species richness, while tussock this data, we determined the native and area was positively correlated to total plant nonnative plant richness for each tussock. species richness (Table 1a). Non-native species richness was 2.3 Statistical Analysis associated with both native plant species richness and one environmental variable, All statistical analyses were conducted distance to upstream tussocks. In the best with JMP statistical software v14. We used fit multiple linear regression model for a multiple linear regression model to native plant species richness, 35% of the analyze how our environmental variables non-native species richness was explained and native plant richness impact non-native by the model (P < 0.0001; Table 1b). Non- plant richness. A log transformation was native species richness was both positively applied to the distributions of four associated with native plant richness and environmental variables (distance to shore, distance to upstream tussocks (Table 1b). distance to upstream tussocks, stream Lastly, invader richness was associated velocity, and tussock area) before analysis with tussock native species richness. There so that they would fit a normal distribution. was a positive correlation between the We defined a full model with our five native and non-native species richness measured environmental variables and (N=126, R2 = 0.28, P < 0.0001; Figure 3). native species richness to perform model selection, choosing the best model based on the lowest AICc values through the “stepwise” function. We then did the same process to investigate native plant richness but ran the model with only five environmental variables. To examine the relationship between native and non-native species richness, we used a linear regression to compare the two variables.

RESULTS Figure 3. Linear regression of number of non-native We found that several measured species by number of native species on tussocks environmental variables were associated formed by Carex nudata (torrent sedge). Results of with native plant species richness. In the the linear regression yielded R2 = 0.28 and P < 0.0001 best fit multiple linear regression model for (N = 126; F = 48.6; DF = 1, 125). native plant species richness, 38% of the native plant species richness was explained by four environmental variables (P <

CEC Research | https://doi.org/10.21973/N3WM1Z Spring 2019 5/9

a. Native species richness (R2 = 0.38 P < 0.0001 N = 118; F = 18.8; DF = 4, 117)

Variable Estimate S.E. P

(1) Log Distance to Upstream -1.40 0.52 0.007 Tussocks

(2) Log tussock area (m2) 2.97 0.48 <0.0001

(3) Log stream velocity (m/s) -1.32 0.48 0.007

(4) Distance along group -1.38 0.60 0.02

b. Non-native species richness (R2 = 0.35 P < 0.001 N = 120; F = 33.1; DF = 2, 119)

Variable Estimate S.E. P

(1) Log Distance to Upstream 0.41 0.11 0.003 Tussocks (m)

Number of native species 0.17 0.02 <0.0001

Table 1. Stepwise (backward deletion) multiple regression of native and non-native species richness on tussocks against environmental variables from tussocks formed by Carex nudata (torrent sedge).

DISCUSSION the idea that that native propagules that come from upstream are important to the Our results found that native species establishment of native species on tussocks. richness was higher for tussocks that (1) When considering stream velocity, tussocks were closer to the front of a group of that experience a lower stream velocity tussocks, (2) had a larger area, (3) were may allow propagules to be more likely to experiencing a lower stream velocity, and deposit because of a lower stream energy. (4) were closer to upstream tussocks (Table This result indicates that physical variation 1a). In using propagule pressure as a within the stream can influence propagule mechanistic framework to explain the pressure on tussocks. In the case of relationship between the observed distance to upstream tussocks, the environmental variables and native species observed result suggests that propagule richness, there are different ways that pressure manifests from other tussocks. propagule pressure could drive the When tussocks are closer to upstream observed patterns. The fact that native tussocks, there is less distance for species richness increases on tussocks with propagules to travel, and therefore making larger areas and that are closer to the front it more likely for propagules to land on of groups suggests that propagule pressure them. On the other hand, we were comes from upstream. This lends support to surprised to find that (5) distance to shore

CEC Research | https://doi.org/10.21973/N3WM1Z Spring 2019 6/9

did not influence native species richness. example, Levine (2000) suggested that the This suggests that there is no direct positive relationship between non-native propagule pressure from the stream shore. and native species richness on tussocks at Future studies investigating the sources of the community level could be due to native propagules on tussocks can confirm physical factors that covary with native whether native species from shore fail to species richness. In our study, we also contribute to tussock native species found there to be a positive correlation richness. When taking into account all of between non-native and native species these results together, there is strong richness on tussocks at the community level support that environmental factors act to as found by Levine (2000) (Figure 3). facilitate propagule pressure (Von Holle However, we also found that invader 2005). Additionally, the results show that success was influenced by the distance to the way propagule pressure acts to upstream tussocks, even when taking into influence native species richness on account the influence of native species tussocks can vary based on where richness (Table 1b). Additionally, distance to propagules are coming from (Simberloff upstream tussocks had an effect on native 2009, Von Holle & Simberloff 2009). species richness (Table 1a). All of these While native species richness was results taken together show that correlated with several environmental environmental variables that covary with variables, non-native species richness was native species richness can directly only influenced by distance to upstream influence invasion success of tussocks on tussocks, with instead there being a positive the South Fork Eel River. Our results relationship (Table 1b). We propose what support our hypothesis in that may have been occurring is that non-native environmental variables can independently species have dispersion methods that allow act alongside native diversity to influence them to disperse further, and hence habitat invasibility (Levine 2000, Luo et al. increase their propagule pressure on 2018, Von Holle 2005). tussocks that are further away from The most widely applicable conclusion upstream tussocks (Rejmánek & Richardson from our study is that future studies 1996). Our results suggest that distance to examining the relationship between upstream tussocks acts to facilitate diversity and invasion in natural systems propagule dispersal methods employed by should look at a wide range of invaders, because they may in general be environmental variables that can make it better dispersers in this system. Future appear that there is an inherent positive studies looking at dispersion patterns correlation between the success of non- between native and non-native species in native invaders and native diversity (Levine this system will provide support for this & D’Antonio, Luo et al. 2019). While the proposed mechanism. diversity/invasibility hypothesis can still The role that environmental variables play hold true (e.g. Levine 2000, Naeem et al. on invader success has been thoroughly 2000), the need to consider environmental investigated and considered (Levine & variables when investigating the D’Antonio 1999; e.g. Luo et al. 2018, Towers native/non-native species relationship in & Dwyer 2018, Von Holle 2005). For other natural communities (e.g. islands,

CEC Research | https://doi.org/10.21973/N3WM1Z Spring 2019 7/9

grasslands, tropical rainforests) are Enserink, M. 1999. Biological invaders sweep in. necessary to understand the relative Science 285:1834–1836. strength of both abiotic and biotic variables Fridley, J. D., Stachowicz, J. J., Naeem, S., Sax, D. F., on invader success (Belote et al. 2008, Seabloom, E. W., Smith, M. D. & Holle, B. V. 2007. Mazia et al. 2019, Peters et al. 2006, The invasion paradox: reconciling pattern and Simberloff 2009). Applying this framework process in species invasions. Ecology 88:3–17. to all systems can help determine the Holle, B. V., & Simberloff, D. 2005. Ecological relative contribution that diversity and resistance to biological invasion overwhelmed by environmental variables play in resisting propagule pressure. Ecology 86:3212–3218. invasion, thereby allowing conservation managers to better focus resources to Keeley, J. E., Lubin, D., & Fotheringham, C. J. 2003. prevent the establishment of invasive Fire and grazing impacts on plant diversity and alien plant invasions in the southern Sierra species and consequently preserve Nevada. Ecological Applications 13:1355–1374. biodiversity and ecosystem functions. Knops, J. M., Tilman, D., Haddad, D. N., Naeem, S., ACKNOWLEDGMENTS Mitchell, C. E., Haarstad, J. and Groth, J. 1999. Effects of plant species richness on invasion We would like to thank our instructors, dynamics, disease outbreaks, insect abundances Tim Miller and Krikor Andonian, for their and diversity. Ecology Letters 2:286–293. extremely helpful input and advice, as well Levine, J. M. 2000. Species diversity and biological as Ana Miller-Te Kui. We would also like to invasions: relating local process to community thank the steward, Peter Steel, for his pattern. Science 288:852–854. hospitality at Angelo Coast Range Reserve. This work was performed at the Levine, J. M., & D’Antonio, C. M. 1999. Elton revisited: a review of evidence linking diversity University of California’s Angelo Coast and invasibility. Oikos 1:15–26. Range Reserve, doi:10.21973/N3R94R. Levine, J. M., Vila, M., Antonio, C. M. D., Dukes, J. S., REFERENCES Grigulis, K., & Lavorel, S. 2003. Mechanisms underlying the impacts of exotic plant invasions. Belote, R. T., Jones, R. H., Hood, S. M., & Wender, B. Proceedings of the Royal Society of London. Series W. 2008. Diversity–invasibility across an B: Biological Sciences 270:775–781. experimental disturbance gradient in Appalachian forests. Ecology 89:183–192. Luo, Z., Chen, X., Xia, G., & Chen, X. 2018. Extrinsic environmental factors, not resident diversity itself, Case, T. J. 1990. Invasion resistance arises in strongly lead to invasion of Ageratum conyzoides L. in interacting species-rich model competition diverse communities. Ecological Research communities. Proceedings of the National 33:1245–1253. Academy of Sciences 87:9610–9614. MacArthur R.H., Wilson E.O. 1967. The theory of Colautti, R. I., Grigorovich, I. A., & MacIsaac, H. J. island biogeography. Princeton: Princeton Univ. 2006. Propagule pressure: a null model for Press. biological invasions. Biological Invasions 8:1023– 1037. MacArthur, R.H. 1970. Species packing and competitive equilibrium for many species. Elton, C. S. 1958. The ecology of invasions by plants Theoretical population biology 1:1–11. and animals. University of Chicago Press.

CEC Research | https://doi.org/10.21973/N3WM1Z Spring 2019 8/9

Mazía, N., Chaneton, E. J., & Ghersa, C. M. 2019. plant species invade hot spots of native plant Disturbance types, herbaceous composition, and diversity. Ecological Monographs 69:25–46. rainfall season determine exotic tree invasion in novel grassland. Biological Invasions 21:1351– Symstad, A. J. 2000. A test of the effects of 1363. functional group richness and composition on grassland invasibility. Ecology 81:99–109. Naeem, S., Knops, J. M., Tilman, D., Howe, K. M., Kennedy, T., & Gale, S. 2000. Plant diversity Tilman, D. 1999. The ecological consequences of increases resistance to invasion in the absence of changes in biodiversity: a search for general covarying extrinsic factors. Oikos 91:97–108. principles. Ecology 80:1455–1474.

Pejchar, L., & Mooney, H. A. 2009. Invasive species, Towers, I. R., & Dwyer, J. M. 2018. Regional climate ecosystem services and human well-being. Trends and local-scale biotic acceptance explain native– in Ecology & Evolution 24:497–504. exotic richness relationships in Australian annual plant communities. Proceedings of the Royal Peters, D. P., Yao, J., & Gosz, J. R. 2006. Woody plant Society B: Biological Sciences 285. invasion at a semi-arid/arid transition zone: importance of ecosystem type to colonization and Von Holle, B. V. 2005. Biotic resistance to invader patch expansion. Journal of Vegetation Science 17: establishment of a southern Appalachian plant 389–396. community is determined by environmental conditions. Journal of Ecology 93:16–26. Planty-Tabacchi, A. M., Tabacchi, E., Naiman, R. J., Deferrari, C., & Decamps, H. 1996. Invasibility of Holle, B. V., & Simberloff, D. 2005. Ecological species-rich communities in riparian zones. resistance to biological invasion overwhelmed by Conservation Biology 10:598–607. propagule pressure. Ecology 86:3212–3218.

Rejmánek, M., & Richardson, D. M. 1996. What Whitfeld, T. J., Lasky, J. R., Damas, K., Sosanika, G., attributes make some plant species more Molem, K., & Montgomery, R. A. 2014. Species invasive? Ecology 77: 1655–1661. richness, forest structure, and functional diversity during succession in the New Guinea lowlands. Simberloff, D. 2009. The role of propagule pressure Biotropica 46:538–548. in biological invasions. Annual Review of Ecology, Evolution, and Systematics 40:81–102. Wiser, S. K., Allen, R. B., Clinton, P. W., & Platt, K. H. 1998. Community structure and forest invasion by Stohlgren, T. J., Binkley, D., Chong, G. W., Kalkhan, an exotic herb over 23 years. Ecology 79:2071– M. A., Schell, L. D., Bull, K. A. & Son, Y. 1999. Exotic 2081.

CEC Research | https://doi.org/10.21973/N3WM1Z Spring 2019 9/9