Human Influenza Pandemic? The value of understanding the factors that effect disease dynamics

Infectious diseases are a well-studied public health issue. However, the factors that influence infectious disease dynamics that involve elements of population ecology are just as important to study. Spatial and geographical patterns play a large role in the spread, transmission, and elimination of infectious diseases.  For example, understanding human population density gives insight to the rate of which an infectious disease will spread. Also studying the patterns presented in previous occurrences of infectious diseases are a key element in preventing future outbreaks of these diseases. This understanding of density-dependence, which is population growth dependent on the average number of individuals per unit area, also accounts for the persistence and transmission dynamics of infectious diseases. Studying the patterns presented in previous occurrences of infectious diseases are a key element in preventing future outbreaks of these diseases.

Figure: Infectious diseases that affect human populations can be spread through a multitude of animal hosts

Chowell et al. studied the spatial variations (regions and counties) in transmissibility and mortality of the 1918-1919 H1N1 influenza pandemic in rural and urban areas of England and Wales (2008). The authors collected data from the British Ministry of health about weekly influenza-related deaths. Decennial census data provided the demographic information needed to evaluate the effect of the socio-demographic factors-population size and density, residential crowding and urbanization-that led to differential rates in transmissibility (reproductive number), and mortality rural and urban. This study looked at these factors from the autumn of 1918 compared to the winter of 1919. The rate of influenza transmissibility was the highest and also the most statistically significant during autumn 1918 in both urban and rural areas while transmissibility was low and not significant during the winter of 1919. From the socio-demographic factors analyzed, population size and urbanization were found to be significant predictors of the death rate during both times frames of the influenza pandemic. There was a much higher experienced per capita death rate in urban areas during both autumn and winter periods while rural areas with the smallest population sizes experienced the highest death rate overall. Population size was also a significant predictor of the onset of the disease since early onset of influenza was found in areas with larger population sizes.   

Figure 1: Weekly number of influenza during 1918-1919 pandemic at spatial scales of regions (administrative units) and counties

          This paper addresses important aspects of spatial variations found within a disease epidemic while leaving some questions unanswered. For example, the effects of population density on transmissibility and mortality rates were left unexplained. The Chowell et al. study does not completely compare the factors that differ between rural and urban areas that could have led to these spatial differences except for population size. In a more recent paper, the influenza pandemic of 1918 was studied in England, Wales and the US further exploring the factors that led to the spatial spread of the disease paying attention transmission, population size and distance while specifically analyzing the extent to which density-dependence played a role in the results found without looking at these differences in specifically rural and urban areas (Eggo et al., 2011). I think it would be extremely important to do a qualitative analysis of the factors that influence the transmissibility and mortality of human influenza in rural and urban areas, looking at population density as a major factor. I would conduct this study in Iowa, a state with both rural and urban counties, where I would collect data on reported incidences of influenza through public health records and analyze density through census data. Once this data is collected, I would conduct analyses using statistical tests, spatial tests, and general model testing to  truly quantify my results. The results of this study would add to the knowledge of the more recent human influenza dynamics, while examining the importance of population type that affect the spread of disease, access to preventative measure, etc. 

Works Cited

Chowell, G., L. Bettencourt, N. Johnson, W. Alonso, and C. Viboud. 2008. The 1918-1919 

Influenza Pandemic in England and Wales: Spatial Patterns in Transmissibility and Mortality Impact. Proceedings: Biological Sciences, 275:501-509.

Eggo, R. M., S. Cauchemez, and N. M. Ferguson. 2011 Spatial dynamics of the 1918 influenza pandemic in England, Wales and the United States. Journal of The Royal Society Interface. 8:233-243.

Some people just want to watch the world burn: how fire ecology informs land management

                While we see fire as a destructive force, fire is a necessary part of many environments and ecosystems. Fire often drives plant diversity, is responsible for the germination of plant seeds, clears away debris, and puts nutrients and minerals back into the soil. As such, understanding the role of fire disturbance in natural landscape dynamics is essential for park and land management. Fire dynamics, however, are not completely understood (Bergeron et al. 2002.) Moreover, the effect of prescribed burns varies according to a variety of factors.

                Kennard and Gholz (2001) investigated the effect of fire intensity on soil quality and nutrient availability in a Bolivian forest. Their study aimed to test the difference between high and low intensity fires, as determined by “fuel loads” or the amount of woody debris present, in terms of: 1.) how they affected the soil itself and 2.) how it affected the growth of a particular plant, Anadenathera colubrine. They conducted their burns in plots and looked at soil pH, organic matter concentrations, and mineral concentrations (such as phosphorus, nitrogen, and magnesium).

An experiment burn conducted at Kruger National Park.
Image from http://www.fire.uni-freiburg.de/iffn/country/za/za_6.htm

                The study found that high intensity fires led to significantly higher mineral concentrations that the other treatments, but they led to a significantly lower concentration of organic matter in the soil. The amount of organic matter in the soil affects how well soil accepts and holds water. Less soil organic matter suggests that the soil is now worse at accepting and retaining water, and also can lead to greater water evaporation from soil and surface run-off. Meanwhile, while the low-intensity burns also led to higher mineral concentrations than the control plots, they did not decrease soil organic matter. As for the effect of burn intensity on the growth of A. colubrine, the seedlings grew the tallest after high intensity burns; however, as Kennard and Gholz (2001) note, the effects of the high intensity burns may not last in the long term. The soil structure will takes years to recover after these high intensity burns, so the researchers conclude that the low intensity burns may be the safer, more beneficial route.

                While experiments and research into the effect of burn intensity are rare, they may be incredibly useful to consider when developing fire regimes. A burn intensity experiment is on-going in South Africa’s Kruger National Park, where large patches of the landscape have been burned twice at high or low intensities, this time determined by the time of year and season, over the span of five years. These experimental burns were ultimately conducted to inform future fire regimes in the savannas of KNP, fire driven systems. For my proposed research, I am interested in looking at how plant diversity, specifically growth under large trees, differs between the high and low intensity burn sites. Kruger National Park’s experimental burns also have to address the issue a declining population of large trees due to an increasing population of elephants. This research will help explore fire – herbivory interactions, and which type of fire may be more beneficial in promoting greater plant diversity. Moreover, it has the added conservation impact, as developing fire regimes that will cause the least damage to already vulnerable large trees is in the best interest of the savanna ecosystem. The Kennard and Gholz (2001) study is a useful study to consider because it looks at the biogeochemical aspects of prescribed fire intensity. Similar to their predictions and conclusions, I believe that the low intensity burns will ultimately yield the most diversity while also being the least damaging.

Works Cited

Kennard, D. K., and H. L. Gholz. 2001. Effects of high- and low-intensity fires on soil properties and plant growth in a Bolivian dry forest. Plant and Soil 234: 119 – 129.

Bergeron, Y., A. Leduc, B. D. Harvey, and S. Gauthier. 2002.Natural Fire Regime: A guide for sustainable of the Canadian boreal forest. 

When landscape modification is advantageous for protected species. The case of a synanthropic tarantula, Brachypelma vagans

Research by Salima Machkour-M’Rabet, Yann Hénaut, Sophie Calmé, and Luc Legal

Post by Charlotte Kanzler

            Habitat fragmentation and its effects on wild species is a problem which ecologists must confront in an increasingly urban world. Fragmentation occurs when development or another disturbance splits a piece of habitat, decreasing the amount of suitable habitat and increasing the distance (isolation) between those useable patches. This can make it more difficult for a species to disperse from one patch to another, and, as consequence, decreases the genetic diversity of the population, which may suffer from inbreeding. In spite of many theoretical reasons why habitat fragmentation is harmful to a population, empirical evidence reveals a more complicated picture, with some research bearing out these negative effects, while other studies show a neutral or positive effect on biodiversity. The inconsistency of these results make it valuable to study the effects of habitat fragmentation under a variety of conditions and in a variety of species.

            Though fragmentation has some relevance to many species, the researchers in the article in question focused on tarantulas: specifically, certain species in the South American tarantula genus Brachypelma. This genus is at risk because of their members’ limited distribution, high juvenile mortality rates, and long maturation time (7-8 years in males, 9-10 years in females), as well as their high value in the pet trade. These factors combined to make Brachypelma an endangered genus, potentially more vulnerable to fragmentation than other tarantula species.

In the article under question, researchers wanted to figure out if habitat fragmentation was jeopardizing populations of Brachypelma vagans. To do so, Machkour-M’Rabet and contemporaries captured thirty different tarantulas each from the open central plazas of six different villages in the Mexican Caribbean. Five of the populations were continental; the one isolated island population, nestled on Cozumel island, was traced back to a release of tarantulas used in a movie production after filming in 1971.

The region in the Yucatan Peninsula from which Brachypelma vagans were drawn for this study. Note the population isolated on Cozumel Island. Image from Machkour-M’Rabet et al., 2011.

 The plazas on which the tarantulas were collected were oases of suitable habitat surrounded by unsuitable forest, agricultural field, and human development. A piece of the limb was taken from each tarantula (not ideal, but they grow back) in order to perform genetic analysis. Researchers were seeking the presence or absence of loci which could be used to determine the amount of gene flow and the genetic differentiation between sampled populations. Contrary to expectations, the genetic diversity within each population was high, indicating that adults were still able to travel between populations. This movement offset processes such as genetic drift which, reduce genetic variability. Though the island population had less diversity than the others, it was still within acceptable ranges, evidencing its recent isolation.

The results of the Machkour-M’Rabet study indicated that the tarantulas existed in a metapopulation, a population comprised of populations connected by dispersal events - dispersal that, instead of being limited by human development, may have been aided by it (for example, tarantulas can move more easily along gently-used dirt roads cutting through a forest). They assumed that dispersal events were fueled mainly by the males of the species, which range out to find solitary females shortly after they reach maturity.

A typical tarantula burrow, of the species Aphonopelma iodius. Image from http://wildaboututah.org/images/pitts.2004.05.18zc_51804b_DSCN4563.jpg.

           

            Though they assume that dispersal was caused by the males of the species, the authors of the study never explicitly linked male dispersal, habitat fragmentation, and genetic diversity - i.e., they never actually tracked the males during breeding season. This leaves open the question of how important the dispersing males are to the persistence of each population, genetic or otherwise. Though the population may not suffer from genetic homogeneity, if too many males die while migrating, the populations may still be at risk. My grant proposal revolves around some of the questions left unanswered in this report, as well as in the realm of tarantula research at large. Specifically, I’d like to take a closer look at the effect of habitat fragmentation on male mating dispersal in a common desert tarantula, Aphonopelma chalcodes. However, I will assess my results through the lens of population ecology: how differential mortality between sexes could impact population persistence. The effect of sex-based dispersal on metapopulations is not only an understudied aspect of tarantula ecology, but of metapopulation dynamics as a whole: another piece in the puzzle of population dynamics in the face of human intervention.

           

Reference:

Machkour-M’Rabet, S., Hénaut, Y., Calmé, S., Legal, L.2011. When landscape modification is advantageous for protected species. The case of a synanthropic tarantula, Brachypelma vagans. Journal of Insect Conservation 16:479-488.

Citizen Scientists Help Us Understand Plant Diseases

          When trying to elucidate infectious disease dynamics, it is necessary understand spatial and temporal patterns of the species being infected as well as the species that is infectious. When studying plant infectious disease, this means that it is necessary to have robust and up to date data on where the plant is located, and geographically where the plants are being infected. If the disease has a broad range, then the task of gathering up-to-date, robust data is extremely difficult for scientists on their own. Additionally, most data collected by scientists comes from plants on public lands, which leaves a gap in understanding plant diseases on private lands.  

            Meentemeyer et al. recognized citizen science projects as a means to increase data sets and improve knowledge of Sudden Oak Death (SOD), a disease caused by Phytophthora ramorum. In 2008, Meentemeyer et al. founded Sudden Oak Death Blitz, a community science program in California that trains community members to collect leaf samples from oak trees infected with SOD. Each volunteer is able to send in 16 leaf samples, and the location of each sample site is tracked through GPS within a mobile application. Figure one shows the data available for geographic distribution of SOD before and after SOD Blitz. The presence or absence of P. ramorum is verified in a laboratory through Polymerase Chain Reaction (PCR) and ribosomal DNA sequencing. From 2008 to 2013, SOD Blitz had over 1,600 community science volunteers. Prior to the program, they were able to predict P. ramorum infections with 65% accuracy (2008). In 2013, after five years of the program, they were able to predict with 78% accuracy. Additionally, for each year of the program, community science volunteers have had an equal or higher ratio of infected leaves: non infected leaves than self-reported expert volunteers. This indicates that with some training, community scientists can collect samples as effectively as self-reported experts.

Figure 1: On the left are SOD sampling sites pre-SOD Blitz program. On the right are SOD sampling sites five years into the program (2013). Red dots represent infected samples, while green squares represent uninfected samples. 

            The success of SOD Blitz suggests that similar community monitoring programs could be useful in helping scientists understand other plant diseases. One potential opportunity would be to implement a similar program to study bur oak blight (BOB) in the Midwest. BOB is leaf spot disease caused by fungus, Tubakia iowensis which occurs primarily in Iowa and Minnesota (figure 2). Harrington et al. have suggested that the recent increase in BOB may be related to the increase in spring rains in the Midwest. With climate change, it is predicted that Iowa will have increased spring rainfall, so it would be interesting to further investigate the connection between rainfall and BOB disease progression.  Harrington et al. have performed some observations on BOB’s distribution (figure 2), but a community science monitoring program would help expand the data set and monitor disease progression over time in an efficient manner. By creating a long-term program similar to SOD Blitz, we would be able to better understand the dynamics of T. iowensis infections in bur oaks. Additionally, a community science approach could help to engage the public in the process of science and instill an appreciation for the envrionment. 

Figure 2: Geographic distribution of T. iowensis. Red represents counties where T. iowensis presence has been confirmed. Gray represents counties where T. iowensis has not been found as of July 2016.

References:

Meentemeyer, R.K.; Dorning, M.A.; Vogler, J.B.; Schmidt, D.; and Garbelotto, M. 2015. Citizen Science Helps Predict Risk of Emerging Infectious Disease. Frontiers in Ecology and the Environment. 13: 189–194.

What can ants tell us about restoration?

            As human worldwide human activities continue to develop natural lands for human use, conservation programs to conserve natural areas and mitigate development through restoration efforts are becoming increasingly common. In order to evaluate the success of these restoration efforts, scientists have applied a wide variety of techniques to measure the diversity of species occurring in restored areas. Many of these techniques focus on the use of bioindicators – certain organisms whose presence is highly correlated to the quality of the restored habitat. These species – or groups of closely related species – tend to require that specific natural resources are present in their environment, and, if restoration actions successfully increase the amount of these resources that are available, then the bioindicator will also become increasingly common. By focusing on organisms known to be useful bioindicators, studies surveying restored areas can reduce the effort required to determine the success of these restoration projects, allowing more efficient and informative monitoring programs.

Figure 1. Camponotus americanus, a member of Subordinate Camponotini, and a species that might occur in similarly conducted samples in the American Midwest. Photo from http://bugguide.net/node/view/328928/bgpage

            Previous studies throughout tropical rainforest and arid regions in northern Australia have found that ants are one such group of bioindicator; Gollan et al. (2011) show that these same methods can also be applied to temperate regions as well. In this study, the researchers selected 12 sites from within a temperate river catchment within a grassland area of southeast Australia. The chosen sites differed in the amount and type of vegetation present, and they were classified into four classes: mature woodlands, older revegetation (7-10 years post-restoration), young revegetation (less than 3 years post-restoration), and unplanted grassland. In each site, the researchers sampled insects using pitfall traps, designed to contain and collect any insects that stumble into them. After classifying the sampled ants according to morphospecies, which group together difficult to distinguish species with similar ecological functions, the researchers compared the number of species present in each of the four classes of site vegetation. Though they identified more than 20,000 ants from 68 different species, Gollan et al. (2011) found that the collection of ant species present at a site varied more within each vegetation class than between sites of different vegetation classes, showing that analyzing ant communities as a whole does not provide much information on the quality of the restored area. However, when they limited their analyses to certain functional groups – groups of morphospecies lumped together by shared interspecific interactions and food sources – they found that different functional groups responded in vastly different ways to habitat quality. In particular, one group known as Subordinate Camponotini, was common in mature woodlands, moderately present in revegetated areas, and absent in unplanted grasslands, as it likely responds to the availability of its food sources in the different types of vegetation.

            Despite being shown to be valid bioindicators in a wide range of habitats, analyses of habitat quality through measurement of ant communities has seen limited use in the United States. In proposed research, I am interested in exploring the applicability of this system in riparian areas here in the Midwest. In addition to the effects of agriculture, many rivers in the Midwest have been channelized, and these various control methods have resulted in substantial changes to the flooding patterns these rivers have experienced. As periodic disturbance by flooding is key to maintaining the health of riparian areas, human-induced modifications to flooding history may have had widespread impact on the communities of these riparian areas. In my proposed project, I will sample ants in riparian areas with different flooding histories and compare these communities using the methods of Gollan et al (2011). By looking at diversity of different functional groups, as well as overall diversity, I aim to identify the effects that changes to the flooding regime of rivers in the American Midwest has on the quality of adjacent riparian areas.

Reference:

Gollan, J., L. Lobry de Bruyn, N. Reid, D. Smith, & L. Wilkie. 2011. Can ants be used as ecological indicators of restoration progress in dynamic environments? A case study in a revegetated riparian zone. Ecological Indicators 11: 1517-1525. 

Turtles in Suburbia: Habitat Usage in Suburban Retention Ponds

            In recent centuries, human activity has transformed Earth into something almost unrecognizable. Many of the biomes and wildernesses that once covered the planet barely exist anymore – enough so that some scientists suggest we create a new system of classification based primarily on how humans have shaped their environments, rather than on what organisms might theoretically occupy a given region had human development not occurred (Ellis & Ramankutty, 2008). Whether or not such systems end up becoming widely used, however, it is clear that human development is something that ecologists, if they want to understand how populations of organisms work today, need to deal with properly. Fortunately, many have taken up the call: studies of urbanization’s effects on plants and animals are now commonplace. Less common, however, are suburban studies that treat suburban areas as more than an intermediate between “city” and “wilderness” – even though an increasing amount of the U.S. population in particular resides in suburban areas. Recently, however, ecologists have started to look at the suburbs from a different perspective, examining the habitat features unique to these areas and how plants and animals interact with them

.

Figure 1: Location of the study site within the Australian Capital Territory (ACT).

            One particularly insightful perspective into these specifically suburban habitats is “Suburbs: Dangers or Drought Refugia for Freshwater Turtle Populations?” (Roe, Rees, & Georges, 2011). The authors hoped to study whether suburban retention ponds in Canberra, Australia provided lower or higher-quality habitat than nearby nature preserves for eastern long-necked turtles (Chelodina longicollis). They determined this by trapping and tagging turtles over the course of a year, measuring recapture rates and growth rates of recaptured individuals over time, and by taking size and water quality measurements of the bodies of water in the study areas. Despite initial suspicions that these retention ponds might be detrimental to the turtle populations – luring turtles in but then harming them with poor water-quality and low food abundance – the study actually found that turtles in the suburbs grew faster and larger than those in the nearby nature preserves. This suggests that, particularly during times of drought such as the year during which this study occurred, the humble suburban retention pond might serve as a valuable refuge for local wildlife, and should be managed with this ecological function in mind.

Figure 2: Growth rate (y-axis) vs. initial carapace length (x-axis) for turtles in the suburbs (dark circles, solid line) and nature preserves (white circles, dashed line).

            Inspired in part by this study, I intend to carry out my own research on distribution and movement of muskrats (Ondatra zibethicus) in suburban water retention ponds. Although these particular researchers were not certain as to what factors might make the suburbs a better habitat for turtles than the nature preserves – none of the water quality or pond size metrics that they recorded seemed to suggest an explanation – their findings nonetheless suggest that these may be more than just universally “low-quality” habitats. What applies to turtles, however, may not necessarily apply to many of the other species that thrive in these developed areas. By studying how muskrats – another species commonly found in the suburbs, albeit in the United States rather than in Australia – interact with these anthropogenic habitats as opposed to those in more “natural” areas, I hope to build off the research done by Roe, Rees, and Georges in order to further develop our understanding of suburbia not just as an awkward in-between of urban and rural zones, but as a kind of human-made biome that needs to be understood (and managed) as such.

Works Cited

Ellis, E.C., & Ramankutty, N. (2008). Putting people in the map: Anthropogenic biomes of the world. Frontiers in Ecology & the Environment, 6(8), 439-447.

Roe, J.H., Rees, M., & Georges, A. (2011). Suburbs: Danger or drought refugia for freshwater turtle populations? The Journal of Wildlife Management, 75(7), 1544-1552. Retrieved from http://www.jstor.org/stable/41418195. 

Watch Your Step: Human Disturbances to Plant Assemblages

Human activities often cause disturbance to the environment, even when their activity is meant to conserve the environment. In order to manage environmentally protected areas, people usually look for ways to get around. When humans walk around, they risk stepping on vulnerable plants and animals. Furthermore, land managers often use all-terrain vehicles (ATVs) to transport heavy materials long distances. ATVs and foot traffic potentially cause stem breakage and fatalities to surrounding plant species. In addition to land management, humans often walk around or ride ATVs in preserved areas for recreation. Because of the great potential for humans’ well-intending activities to be harmful, we should study the extent of our disturbances on other species.

An experimental study looked into the effect of the land management practice of using all-terrain vehicles to apply mosquito treatments to wetlands. This study looked at the impact of two different types of ATVs at two different intensities on stem breakage and the canopy height of a San Francisco Bay wetland plant, pickleweed (Figure 1).

Figure 1. Salicornia virginica L., the variety of pickleweed featured in the wetland study. Picture from the Digital Atlas of Virginia Flora.

They found that heavier ATVs create more stem breakage and reduce the canopy height. Additionally, the researchers found that the more that ATV ran over a plot, the more time it took for that plot to return to the canopy and stem breakage level before the ATV disturbance. The researchers suggested that ATV usage should be eliminated or reduced and used on a different path each time to avoid damage in coastal marsh systems.

I am proposing to do a similar study at Conard Environmental Research Area, a restored prairie site in Kellog, Iowa. At this site, ATVs are used to spray poison on invasive species, among other reasons. Additionally, most management practices, environmental research, and recreational activities at CERA are done by foot (Figure 2).

Figure 2. Students conduct a research project while walking through the Conard Environmental Research Area. Photo from the Grinnell College website.

My proposal includes setting up experimental plots that will either be ran over by ATVs or walked over by students and local participants. I will be looking at the effect of foot traffic and ATV usage on diversity, stem breakage, and canopy height on the surrounding plant assemblages. Additionally, I will study how these disturbances affect the soil compaction and how long it takes for the plant assemblages and soil to recover to a pre-disturbance state, if at all. This study will intend to scale up research on ATVs that looks at only a single species and also fill in a research gap for the prairie setting. Additionally, this project will engage the public to educate others about the effects of human recreation and management on the environment.

References

Groom, J.D., L.B. McKinney, L.C. Ball, and C.S. Winchell. 2007. Quantifying off-highway vehicle impacts on density and survival of a threatened dune endemic plant. Biological Conservation 135: 119-134.

Hannaford, M.J. and V.H. Resh. 1999. Impact of all-terrain vehicles (ATVs) on pickleweed (Salicornia virginica L.) in a San Francisco Bay wetland. Wetlands Ecology and Management 7: 225-233.

Schlater, T., A., L. Thompson, and S. Price. 2007. Vehicles versus conservation of invertebrates on sandy beaches: Mortalities inflicted by off-road vehicles on ghost crabs. Marine Ecology 28: 354-367.

Which niche is which: how to best use niches to model and predict species distribution

The ecological niche is a foundational idea in ecology, and the vigor of debate around how to best understand and visualize the niche is second only to that of disagreements about how the word is pronounced. The concept of a niche is closely related to the dynamics of distribution, and for many reasons the actual distribution of a species will often not exactly match the distribution made possible by the species' physiological requirements.  Accurate descriptions, predictions, and models of a species’ niche can be hard to come by because the specifics of how the ecological niche functions are so varied. However, when niches are modeled well, they can provide information not only on the role and requirements of a species in its current ecosystem, but also how the species will respond to larger changes in environmental conditions over time. The aquatic ecosystem in particular is home to several species which are well suited to a variety of environmental conditions, which makes them excellent local adapters but also well-suited to become invasive.

Kriticos and Brunwell’s 2016 paper “Assessing and Managing the Current and

 Future Pest Risk from Water Hyacinth, (Eichhornia crassipes), an Invasive Aquatic Plant Threatening the Environment and Water Security” uses niche modeling to predict the future habitat of a common invasive aquatic plant. By using CLIMAX, a niche modeling system, they sought to map the current global distribution of water hyacinth and how that distribution would change under the Global Climate Model’s projection of future climate conditions. CLIMAX uses a series of functions fitted to data on water hyacinth population growth rate under heat and cold stress to find the probability of water hyacinth introduction or persistence at a given location.  Climate is one of the most significant limiting factors of growth for water hyacinth, since it is winter hardy but sensitive to frost. Using CLIMAX to model niche possible niches, Kriticos and Brunwell generated the following maps of climate suitability for water hyacinth, both for the current climate (Figure 3) and predicted future climate (Figure 5). Given predicted climate change, heat stress will reduce the potential for water hyacinth to survive in Saharan Africa, the Middle East, and India, but it will likely able to expand its in low altitude areas such as Tasmania and the South Island of New Zealand. Knowing the many instances of suitable habitat it our current climate, the authors suggested a solution to the spreading invasive species in the form of legislation preventing sale and distribution of water hyacinth, as well as a widespread public education campaign about how to protect against it as an invasive species.

Figure 3. Possible distribution of water hyacinth given current climate conditions

Figure 5. Possible distribution of water hyacinth given future climate predictions

As is shown by the wide variety of suitable habitat for this floating aquatic plant, species distribution can be closely tied to physiological tolerances. Modeling niches based on the known growth rates at certain temperatures is a logical way to predict where water hyacinth will be able to spread. However, simulated niche models do not always take into account other environmental factors such as nutrient availability, and ignore the possibility of discrepancies between the fundamental niche of water hyacinth and its realized niche. Research designs which directly compare the physiological limits of an organism and the conditions of the environments they actually occupy provide details of species distribution that are left out by studies which assume growth rate as a result of physical limits as the only determinant in a species niche.  In my proposed research I intend to apply such a design in order to study another aquatic floating plant, Spirodela polyrhizia. S. polyrhizia, like water hyacinth, can also survive a variety of environmental conditions and spread so quickly as to be a problem in ponds and lakes. Learning more about how to best model and predict its niche will help the people who are trying to manage it, as well as add to the growing body of knowledge on the relationship between niche and species distribution.

Kriticos, D., S. Brunnel. 2016. Assessing and Managing the Current and

 Future Pest Risk from Water Hyacinth, (Eichhornia crassipes), an Invasive Aquatic Plant Threatening the Environment and Water Security. PLOS One. 11(8): 1-18.