Andrés J. Andrade – Removal of the shrub Cornus drummondii: An analysis of two restoration techniques for the tallgrass prairie (Mentors: John Blair and David Hartnett).
The tallgrass prairie is one of North America’s most endangered ecosystems with approximately 4% of its historical extent remaining.  This remaining fraction is endangered by various factors, one of the largest being C₃ woody species expansion, converting the tallgrass prairie to forest.  Fire, an integral part of the tallgrass prairie, has been excluded from most of this diminished ecosystem, which has allowed for increased establishment and recruitment of native, early successional C₃ shrub species.  Of particular concern at Konza Prairie Biological Station is Cornus drummondii (rough-leaf dogwood), a rapidly spreading species that multiplies vegetatively to form shrub islands.  Such islands spread radially and can: 1) displace native prairie species at the periphery, 2) shade out any remaining species in the understory, and 3) facilitate conversion of prairie to forest by providing suitable habitat for late successional C₃ arboreal species, which eventually transform shrub islands to forest tracts.  Starting in 2000, two restoration techniques for the removal of C. drummondii were executed in specific watersheds at Konza Prairie: annual shrub island burnings, and mechanical/herbicidal removal.  This study centers upon the post-restoration analysis of both techniques in order to determine the efficacy of each, quantified through Morisita’s index of community composition, species diversity, species richness, total woody cover, and proportion of forb cover.  Based upon a hypothetical sere model detailing the transition from prairie to forest, a total of five communities were chosen for sampling: post-fire (FD), post-mechanical removal (MR), Grassland State (GS) as the desired end point of restoration and positive control group, undisturbed shrub islands (USI) as the negative control, and Shrubland Transition State (ST) as an intermediate control.  While the MR and FD communities were 73% similar, in what they differed the MR communities were greater in similarity to the desired GS communities with 63% similarity, as opposed to the FD communities which were only 54% similar.  Conversely, the FD and USI communities were more similar (36%) than the MR and USI communities (18%).  The MR communities possessed a higher mean species diversity (2.31) and mean species richness (7.2) than the FD communities (1.60 and 4.5, respectively).  The FD communities possessed the second highest total woody cover (20.0) after the USI communities (40.4).  The MR, GS, and ST communities all possessed significantly comparable low woody cover (10.4, 9.4, 4.8, respectively).  There was no significant difference in the proportion of forb cover among all five groups.  After seven years, the data describe the MR communities as having a more similar composition to the desired GS state than the FD communities, and, consequently, mechanical/herbicidal removal of C. drummondii appears to be a more effective restoration technique than fire within the short-term.

Melissa A.R. Blundell – Are brood parasites unique in their begging? A comparative approach with the Red-winged Blackbird (Mentors: Jim Rivers and Brett Sandercock).
Obligate brood parasites lay their eggs in the nests of other individuals and rely on those hosts to raise their young. Kin selection theory predicts that parasitic young should beg more intensely than host young due to the lack of genetic interest in the survival of their nest mates or in the host reproductive success. To test if begging displays are unique in an obligate brood parasite, the Brown-headed Cowbird (Molothrus ater), I compared the begging displays of the cowbird young with that of a closely related non-parasitic relative, the Red-winged Blackbird (Agelaius phoeniceus). I predicted that the cowbird young would have more exaggerated begging displays than the blackbird, and would result in a higher rate of host food provisioning. Single cowbird young and a single blackbird young were video recorded in the nest of a common parasitic host, the Bell’s Vireo (Vireo bellii). Begging posture, call rate, frequency of begging, volume food provisioned to young and visit rate were measured. All variables measured were similar between the cowbird and blackbird. These results suggest that the cowbird young are not uniquely adapted in their begging behavior. This similarity coincides with Molothrus as the youngest clade of brood parasites.  Possibly a more specialized parasitic behavior in the young has not yet evolved as opposed to older parasitic lineages of birds where parasitic young kill host nestlings.

Adam J. Carter – Subsidies of Kansas streams: the diet of an omnivorous fish (Mentors: David Hoeinghaus and Walter Dodds).
Ecosystems have porous boundaries, with organisms, nutrients and energy passing between adjacent systems.  These subsidies may play an important role in structuring the recipient food web, with potentially cascading ecosystem-level effects.  Stream ecosystems interact with the terrestrial landscape through the riparian zone.  For example, insects emerging from the stream may subsidize terrestrial food webs, and insects falling into the stream may subsidize the stream food web.  Even though riparian zones are increasingly being altered or reduced by humans, only limited research has investigated the role of riparian characteristics on the relative importance of terrestrial subsidies to stream food webs.  In this study, we used an omnivorous fish, the Red Shiner (Cyprinella lutrensis) as a study organism to evaluate the effects of percent canopy cover on the relative importance of terrestrial-derived items in its diet.  The red shiner is a model study organism because it is known to consume both terrestrial and aquatic food items, and is widespread and locally abundant in prairie streams.  I hypothesized that as percent canopy cover decreased, red shiners would consume more aquatic food items and fewer terrestrial food items.  Fourteen stream sites were sampled across northeast Kansas, and physical measurements of each stream (depth profiles, substrate composition, percent canopy cover) were taken at five transects spaced approximately 30 meters apart.  Standard seining methods were used and all fish were identified in the lab to the species level.  Gut contents of 30 random red shiners from each site were identified and quantified.  Importance of dietary items was evaluated using the feeding index IA which incorporates the frequency of the diet item as well as the percent volume.  Terrestrial organisms were dominant diet items in more than half of the sites, and all sites had some type of terrestrial food contribution, illustrating the importance of riparian-derived subsidies.  Percent canopy cover was not significantly correlated with the overall dietary importance of combined aquatic food items or combined terrestrial food items across all sites.  However, the dietary importance of some individual prey categories (e.g. algae, Hymenoptera, Coleoptera, Lepidoptera) were related to percent canopy cover.  Further riparian measurements, such as tree density, may improve our understanding of factors affecting the relative importance of terrestrial subsidies to stream food webs.     

Anthony C. Dalisio – The response of male Dickcissels to geographic song variation (Mentors: Tim Parker and Bill Jensen).
In some oscine songbird species, young males imitate the songs of the neighbors they encounter when they settle on their adult territories. This leads to geographic patterns of song sharing, sometimes known as dialects. Male Dickcissels (Spiza americana) exhibit dialects on a microgeographic scale as expected if this process of adult neighbor song learning is at work. Song sharing between male Dickcissel decreases as distance increases up to 10 km between birds. This led us to ask what selection pressures favor learning song from adult neighbors in the Dickcissel. We tested the “Deceptive Mimicry” hypothesis in which selection favors imitation of the songs of established males because by sounding similar to established birds, the mimic will elicit less aggression from established birds. Our experiment tested territorial males’ responses to three different song playback treatments: 1) songs from their own territory (local), 2) songs from nearby neighborhoods (intermediate) and 3) songs from populations far away singing radically different songs (foreign). In playback experiments, males tended to respond very aggressively to the local song playback. During the trials, they sang the most songs to local and foreign playbacks, and after the trials they sang the most songs to local and intermediate playbacks. Further, Dickcissel males made more flights towards the speaker during playback of local songs than during the other treatments, and after the trials, males flew most towards the speaker for the local and foreign songs. These results contradict the prediction of the Deceptive Mimicry hypothesis that local songs would elicit the least aggression. Thus, we expect another selection pressure favors learning song from neighbors.

Jesús E. Gómez-Carrasquillo – Hide-and-seek among wolf spider and grasshoppers: applying ecology of fear theory  to invertebrates (Mentor: Tony Joern).
The past study was realized at Konza Prairie Biological Station at Kansas in which the ecology of fear theory used to describe the nonlethal interactions between major vertebrate carnivores and their prey. The theory was applied to spider (Lycosidae) and grasshoppers (Acrididae) community interactions. The goals of the past study were to determine if spider presence alters the spatial distribution of grasshoppers, to determine variability in grasshopper and spider density among sites, and their relationship to one another. We predicted that grasshoppers will feed more in areas with a lower predation risk, but that vegetation structure would affect the spider and grasshopper community interactions and abundance. The Ecology of Fear predicts that spider abundance should alter grasshopper feeding.  Spider abundance was determined using pitfall traps at nine sites.  Body sizes were measured to the nearest mm and each spider was identified to Family, Genus and Species level. Grasshopper densities were determined with ring count transects, and feeding distributions were measured using bioassays plot inside the spider pitfall trap plots. Vegetation biomass was determined for five 0.2 m² clip plots (forbs & grass). Our results show that there is no relationship between vegetation biomass and spider or grasshopper density. The negative relationship between grasshopper and spider density suggests that biotic interactions are important. Spatial distribution of feeding by grasshoppers was influenced by spider density. The spatial feeding distribution of grasshoppers within the plot followed predictions of the Ecology of Fear.

Auctavia D. Grant – The effects of grazing on prairie butterflies (Mentor: Jodi Whittier).
Our study examined the grassland butterfly communities found in grazed and ungrazed areas on the Konza Prairie biological station. Bison (Bison bison) was the large ungulate in the grazed units.  We suspected that we would find more adult butterflies using the grazed units because grazing tends to promote increased prevalence of forbs. We aimed to establish the species richness of the butterfly community was higher in the grazed area rather than in an ungrazed area. We established study sites in grazed and ungrazed sites that were all burned on a 4-year cycle. Each site had 4 transects in the uplands and 4 in the lowlands to examine the breadth of species using each treatment. Transects were sampled 3 times over a period of 5 weeks. Along each transects the percent cover of flowering plants was estimated. As expected the total number of butterflies was higher on the grazed sites compared to the ungrazed sites. We saw a similar result for the flowering fords with nearly three times the coverage on the grazed transects. Our results suggest that vertebrate grazing may increase, in cover of flowering plants, and increase species richness of butterflies, as well as a higher taxon diversity in the ecosystem.

Lauren M. Hallett – Effect of warming and altered precipitation of the growth rates of two dominant C₄ grasses (Mentor: Melinda Smith).
Anticipated effects of climate change in the Midwest United States include increased temperatures and more extreme precipitation regimes. Previous research suggests that the two dominant C₄ grasses in North American tall grass prairies, Andropogon gerardii and Sorghastrum nutans, differentially respond to changes in temperature and precipitation. To determine if differences exist at the level of individual growth, relative growth rates of both species were compared across a control and three treatment types: altered precipitation, increased temperatures, and altered precipitation and increased temperatures. The altered precipitation regime was achieved using the Rainfall Manipulation Plots (RaMPs), in which rainout shelters divert, collect, and reapply rainfall to increase the inter-rainfall period by 50% while keeping total amount of precipitation the same. Temperatures were increased by ~2^o C using infrared lamps suspended above the plant canopy. Non-destructive growth measurements (height, number of leaves, leaf area) were taken of 10 representative individuals per species throughout the growing season. Prior to each set of measurements, allometric equations were generated to link non-destructive measurements to dried aboveground biomass by measuring and harvesting individuals in areas adjacent to the RaMPs. R-squared values for these equations were always > 0.9. Consistent with previous studies at the level of physiology and ANPP, Andropogon and Sorghastrum exhibited differential responses to alterations in temperature and precipitation at the level of individual growth.  Andropogon exhibited a higher sensitivity to altered temperature, with consistently higher relative growth rates in the control and delayed-control plots than in those with increased temperature. In contrast, Sorghastrum was more responsive to altered precipitation, with lower relative growth rates consistently observed for the delayed-warmed treatment. Because Andropogon and Sorghastrum are dominant tall grass species, differences in their performance at the individual level may have large scale implications for tall grass prairie ecosystem response to climate change.

Sara L. Jackrel – Evaluation of habitat use by nesting grassland birds and their snake predators (Mentors: Page Klug and Kim With).
The tallgrass prairie is the most endangered ecosystem in the United States, therefore further degradation of this rare habitat impacts many species.  Grassland birds are of particular concern, having declined more than any other group of birds in North America.  Vital to the long term success and recovery of grassland birds is reproductive success.  The primary factor influencing nest failure is predation, with snakes being responsible for many of these predation events.  Despite high rates of nest predation, little is known on how management practices impact snake movement and snake habitat selection.  One concern is that habitat alteration may simultaneously attract both birds and their predators, creating predation hotspots.  I evaluated habitat use of the Eastern Racer and the Great Plains Rat Snake, two common snake species found on KPBS, and compared this with habitat used by nesting grassland birds, predominantly Dickcissels.  Nest monitoring results showed snakes as important nest predators on KPBS, possibly accounting for up to 72% of all nest failures, with a greater rate on four year burns compared to annual burns.  From the fifteen vegetation variables analyzed at nest sites, Program MARK identified high vegetation height matched with low percent live woody coverage as the best candidate model.  Therefore, the higher the vegetation height with less live woody coverage, the higher the nest’s daily survival rate.  A Discriminant Function Analysis, explaining the variance between habitat use of the two snake species as well as fledged nests and predated nests, was most heavily loaded on vegetation height and live wood.  From the separation of the fledged nests from the predated nests shown by the DFA, it can be concluded that the lower the vegetation height and the higher percent live woody coverage surrounding a bird nest, the more likely that nest was predated.  With habitat of Eastern Racers  more similar to that of predated nests and habitat of Great Plains Rat Snakes more similar to that of fledged nests, the Eastern Racer may be the snake species more responsible for nest failure.  Since habitat indirectly impacts nest survival by its direct impact on predator populations and behaviors, knowing the snake community found in regions of the tall grass prairie is important to understand the effects of habitat on these predators.  Since my findings suggest that nest survival is non-random, my research emphasizes the need to provide grassland birds with a type of habitat that provides them with the highest chance of successfully fledging a nest.  In this case, the rate of snake predation may be reduced with management techniques that maximize vegetation height while minimizing invasion of woody species, reducing quality habitat for snakes and other predators and promoting grassland bird population recovery. 

Tyler J. Kohler – Effects of nutrient loading and grazing on periphyton communities in experimental streams (Mentor: Keith Gido).
Prairie Streams are one of the most endangered habitats on the planet, largely due to anthropogenic activities which can lead to excessive nutrient loading.  Nutrient influx can have a significant effect on the biomass, community composition, and stoichiometry of periphyton in these habitats.  Grazing fish can have an impact on the nutrient cycling of these systems due to the trophic level at which they feed, as well by the redistribution of nutrients and altering of nutrient ratios in the water body through excretion.  Therefore, the interaction between these grazers and nutrients may aid in providing us with a better understanding of the dynamics of these systems, as well as have applications for conservation.  This lead us to the development of two questions: 1) How is periphyton biomass, community composition and nutrient ratios affected by gradients of nutrient loading and fish density, and 2) Does the proximity of the periphyton to the rock have an effect on these variables?  In our experiment, we used the experimental streams on Konza Prairie Biological Station to manipulate 6 different nutrient loading treatments, as well as 6 different grazing densities of Southern Redbelly Dace (Phoxinus erythrogaster) to test how these two variables interact to affect periphyton.  Samples were taken from both the riffle and pool portions of the experimental streams, and were further stratified into over and under-story components by removing material at different proximities relative to the substrate.  Samples were then analyzed for biomass, C:N ratios, and community composition.  Our data suggested that increasing nutrient loading had a significant effect by increasing biomass, decreasing the ratio of green algae to diatoms, and decreasing C:N ratios.  Grazers, while having a weak effect on biomass and community composition, were found to have a significant effect by lowering the C:N ratio of over-story periphyton as fish density increased.  The weak effects of the grazers on biomass and community composition is consistent with previous studies performed on Konza Prairie. These results may be due to foraging stimulating or maintaining productivity rather than decreasing it, as well as the possibility that grazers are feeding without preference.  The lowered C:N ratio produced as an effect of higher grazer densities may increase the nutritional value of the periphyton for other organisms, as well as have effects on community structure.

Clairisse E. Nash – Applying a new herbarium database: Geographical patterns of collection of introduced species and targeted plant collecting (Mentor: Carolyn Ferguson).
An herbarium is a natural history museum of preserved plant specimens. The Kansas State University Herbarium (KSC) houses a collection of over 200 thousand specimens, and has one of the largest collections of historical material of agricultural importance in the state of Kansas.  Historical herbarium collections are extremely useful for taxonomic studies in floristic change over time, which aids in ecological forecasting.  Botanists also use herbariums to familiarize themselves with morphological characteristics of particular groups of plants.  Researchers at K State, including KSC Curator C. Ferguson, have done extensive studies tracking the spread of introduced plant species in Kansas and have produced a list of 314 introduced species that are known to exist in Kansas counties (Woods et al. 2005. Sida 21: 1695-1725).  I searched the KSC database and made GIS based maps that show county distribution of collections for 272 of the species listed in the Woods et al.  I also collected and documented plants for use at KSC, some of which had not been previously documented at Konza Prairie. The maps that were made show what counties KSC has collections for each plant as well as when the most recent collection was taken. In the future, the maps can now be used to facilitate targeted plant collecting by trained botanists so that they may update KSC holdings and to also aid in documenting where the plants occur.

Katie Parsons – The response of the dominant C₄ grasses to extreme temperatures (Mentor: Alan Knapp).

Andrea M. Severson – High water habitat: Fish populations in two Kansas River backwater areas (Mentor: Craig Paukert).
The Flood Pulse Concept states that flooding in large rivers may benefit fishes by providing spawning and nursery habitats as well as increased productivity.  The Kansas River flooded in May 2007 and provided an opportunity to test the Flood Pulse Concept.  I hypothesized that the flooding would result in greater fish abundances in the backwaters than in the main channel, and that fish would experience faster growth in the nutrient-rich backwater areas.  Fishes in two Kansas River backwater and adjacent main channel areas were sampled using electrofishing and seining.  No differences were found in fish abundances in the main channel or backwater areas, which did not support the hypothesis that the backwater areas would have greater fish abundances.  However, two small-bodied fish species in one of the backwaters averaged greater lengths than the same species sampled in the adjacent main channel.  These data appear to support the hypothesis that fish would experience faster growth in backwater areas.  However, it is possible that larger fish were the result of increased sampling efficiency or an increased proportion of larger, spawning fish in the backwater versus the main channel.  Also, it was assumed that these fish were all of the same age class, although no aging or determination of hatch date was attempted, and thus it is possible that the larger fish were older than their main channel counterparts.  Nevertheless, backwater areas do appear to be productive, and further sampling in the future may reveal the true effects of the 2007 flooding.

Last updated: November 2007