Konza Prairie Long-Term Ecological Research Research Articles

Flux and stable isotope fractionation of CO2 in a mesic prairie headwater stream

Abstract The carbon dioxide (CO2) fluxes from headwater streams are not well quantified and could be a source of significant carbon, particularly in systems underlain by carbonate lithology. Also, the sensitivity of carbonate systems to changes in temperature will make these fluxes even more significant as climate changes. This study quantifies small-scale CO2 efflux and estimates annual CO2 emission from a headwater stream at the Konza Prairie Long-Term Ecological Research Site and Biological Station (Konza), in a complex terrain of horizontal, alternating limestones and shales with small-scale karst features. CO2 effluxes ranged from 2.2 to 214 g CO2 m−2 day−1 (mean: 20.9 CO2 m−2 day−1). Downstream of point groundwater discharge sources, CO2 efflux decreased, over 2 m, to 3–40% of the point-source flux, while δ13C-CO2 increased, ranging from −9.8 ‰ to −23.2 ‰ V-PDB. The δ13C-CO2 increase was not strictly proportional to the CO2 flux but related to the origin of vadose zone CO2. The high spatial and temporal variability of CO2 efflux from this headwater stream informs those doing similar measurements and those working on upscaling stream data, that local variability should be assessed to estimate the impact of headwater stream CO2 efflux on the global carbon cycle.

White-Tailed Deer (Odocoileus virginianus) Browsing Preferences and Spatial Patterns in a Kansas Flint Hills Riparian Woodland

In many eastern North American forests white-tailed deer (Odocoileus virginianus) populations are large and their browsing reduces tree regeneration and alters sapling species composition. In comparison to forested communities further east, deer effects in Great Plains riparian woodlands are poorly understood. These woodlands may experience greater browsing pressure than regional deer densities suggest because deer are associated with wooded habitats in this otherwise open landscape. To explore potential effects of deer in these riparian woodlands, we quantified deer preference among tree species and spatial patterns in browsing in the Kings Creek riparian zone at Konza Prairie Long-Term Ecological Research Site. Our objectives were to determine how 1) browsing intensity differs among tree species and 2) browsing intensity varies in relation to a woodland edge – center gradient, an upstream – downstream gradient and proximity to deer trails. In June 2016, we quantified the proportion of twigs with new growth that had been browsed on saplings 15 -150 cm tall in 10 m2 plots along pairs of transects that extended from the stream to the woodland edge. Transect pairs were arranged in a downstream progression. We found significant differences among tree species in proportion of twigs browsed with bitternut hickory (Carya cordiformis), black walnut (Juglans nigra) and burr oak (Quercus macrocarpa) being most browsed. Hackberry (Celtis occidentalis) and honey locust (Gleditsia triacanthos) were least browsed. Xerophytic tree species (oaks, hickories) were browsed more than mesophytic species (all other species). Browse intensity was greatest within 3 m of deer trails and increased downstream, but was unaffected by proximity to woodland edges. Strong differences among tree species in browse intensity suggest the potential for deer to affect tree species composition, possibly accelerating mesophication. Even in these wooded habitats, deer browsing is concentrated near deer trails.

Effects of Changing Meteoric Precipitation Patterns on Groundwater Temperature in Karst Environments.

Climate predictions indicate that precipitation patterns will change and average air temperatures will increase across much of the planet. These changes will alter surface water and groundwater temperatures which can significantly affect the local and regional environment. Here, we examine the role of precipitation timing in changes to groundwater temperature in carbonate-karst aquifers using measured groundwater level and temperature data from the Konza Prairie Long-Term Ecological Research Site, Kansas. We demonstrate that shifts to increased cool-season precipitation may mitigate the increases in groundwater temperature produced by increases in average annual air temperature. In karst, the solution-enlarged conduits allow faster and focused recharge, and the recharge-event temperature can strongly influence the groundwater temperature in the aquifer. Our field data and analysis show that predictions of future groundwater conditions in karst aquifers need to consider changes in precipitation patterns, in addition to changes to average annual air temperature.

Scaling climate change experiments across space and time

Scaling climate change experiments across space and time

The effect of precipitation events on inorganic carbon in soil and shallow groundwater, Konza Prairie LTER Site, NE Kansas, USA

Abstract Monthly sampling for 1 year at the Konza Prairie LTER (Long-Term Ecological Research) Site in northeastern Kansas shows a connection between the annual cycles of CO2 in soil air and shallow groundwater DIC (dissolved inorganic C). Soil air CO2 reached 6–7% in July to mid-August, when moisture was not limiting to soil respiration. Following the annual maximum there was a sequential decrease in CO2 in three soil horizons to less than 0.5% because of moisture deficiency in the late summer and temperature decline in the fall and winter. Groundwater pCO2 reached its maximum of 5% in October; the lag-time of 2–3 months may correspond to the travel time of soil-generated CO2 to the water table. The time-variable CO2 caused an annual carbonate-mineral saturation cycle, intensifying limestone dissolution and DIC production when CO2 was high. The C flux depended on respiration and rainfall regimes, and had two main pathways. Transport of soil CO2 in the dissolved form with diffuse flow of recharge water was the most effective during the growing season so long as soil moisture was present. Downward movement of gaseous CO2 and equilibration with groundwater at the water table was favorable in July to August. Storm rainfall events recharged the aquifer within a few hours through preferential flow and stream–groundwater interaction, resulting in dilution of groundwater rather than forcing entrapped CO2 downward. Calculated C flux from the unsaturated zone to the unconfined aquifer in the monitoring period was 0.26 ± 0.03 M/m2/a of C, which is less than 1% of the CO2 that is released by soil to the atmosphere via efflux. However, meteoric precipitation was only 72% of average annual precipitation during the study period, so this study represents dry-condition flux. In addition, increased respiration rates due to warming of the atmosphere have the potential to cause a higher C flux to the saturated zone, intensifying weathering and groundwater acidification, so that further study is suggested.

Quantifying sediment transport across an undisturbed prairie landscape using cesium-137 and high resolution topography

Soil erosion is a global environmental problem, and anthropogenic fallout radionuclides offer a promising tool for describing and quantifying soil redistribution on decadal time scales. To date, applications of radioactive fallout to trace upland sediment transport have been developed primarily on lands disturbed by agriculture, grazing, and logging. Here we use 137Cs to characterize and quantify soil erosion at the Konza Prairie Long-Term Ecological Research (LTER) site, an undisturbed grassland in northeastern Kansas. We report on the small scale (< 10 m) and landscape scale (10 to 1000 m) distribution of fallout 137Cs, and show significant variability in the concentrations and amounts of 137Cs in soils at our site. 137Cs soil concentrations and amounts typically vary by 10% to 30% on small scales, which most likely represents the spatial heterogeneity of the depositional processes. Landscape scale variability of soil 137Cs was significantly higher than small scale variability. Most notably, soils collected on convex (divergent) landforms had 137Cs inventories of 2500 to 3000 Bq m − 2 , which is consistent with the expected atmospheric inputs to the study area during the 1950s and 1960s. Concave landforms, however, had statistically lower inventories of 1800 to 2300 Bq m − 2 . The distribution of 137Cs on this undisturbed landscape contrasts significantly with distributions observed across disturbed sites, which generally have accumulations of radioactive fallout in valley bottoms. Because the upslope contributing area at each sampling point had a significant negative correlation with the soil inventory of 137Cs, we suggest that overland flow in convergent areas dominates soil erosion at Konza on time scales of decades. Very few points on our landscape had 137Cs inventories significantly above that which would be predicted from direct deposition of 137Cs on the soil surface; we conclude therefore that there is little net sediment storage on this undisturbed landscape.

ECOLOGICAL CONSEQUENCES OF C4GRASS INVASION OF A C4GRASSLAND: A DILEMMA FOR MANAGEMENT

Many successful exotic invasive species are functionally distinct from the dominant native species they displace. Occasionally invasion occurs where the exotic species possesses functional traits relatively similar to those of the dominant native. We examined the ecological consequences of such an invasion within a mesic, temperate grassland at the Konza Prairie Long-Term Ecological Research site. We assessed potential changes in carbon (C) and nitrogen (N) cycling and plant diversity following the invasion of a C4 bunch grass species, Andropogon bladhii, into a tallgrass prairie dominated by the native C4 grass species, A. gerardii. In these prairies burning is an important management tool used to maintain native-species dominance. We determined how frequent spring fires affected the impacts of A. bladhii in this system. Over a two-year study our results show that burning regulated the effects that the invasive species has on the native prairie. Compared to the native species, A. bladhii exhibited significantly greater plant biomass, significantly lower pools of soil N, significantly lower rates of decay and C cycling, and higher foliar and root tissue C:N ratio in response to burning. Notable spatial heterogeneity in C and N cycling was evident in areas dominated by the invasive bunch grass. In addition to altered ecosystem processes, areas dominated by the invasive, A. bladhii, had significantly lower plant species diversity. In a grassland ecosystem where burning is an important management tool for controlling exotic-species establishment, maintaining native-species dominance, and increasing productivity, A. bladhii may be able to successfully out-compete the native C4 grass species by using traits typically used to explain the dominance of the native species. With frequent fire, the invasive species has the potential to decrease long-term fertility by lowering N inputs in litter and increasing erosion in non-vegetated soil between bunches, while also having a negative effect on plant diversity. By using fire to promote native C4 grasses and maintain these tallgrass prairies, the threat of invasion by nonnative C4 species may raise a dilemma for future management of these C4 grasslands.

Covariance of biophysical data with digital topographic and land use maps over the FIFE site

Sampling design is critical in locating ground sampling stations for large‐scale climatological field experiments. In the stratified sampling design adopted for the First International Satellite Land Surface Climatology Project (ISLSCP) Field Experiment (FIFE), the study region was stratified into 14 different terrain units based on land use/land cover and topographic variables that were hypothesized to have a strong influence on surface biophysical properties. Digital terrain maps were produced to facilitate ground data integration and extrapolation. This paper describes the biophysical stratification of the FIFE site, implementation of the stratification using geographic information system (GIS) techniques, and validation of the stratification with respect to field measurements of biomass, soil moisture, Bowen ratio (β), and the greenness vegetation index (GVI) derived from thematic mapper satellite data. Maps of burning and topographic position were significantly associated with variation in biomass, GVI, and β. The effects of burning and topography were stronger for the Konza Prairie Long‐Term Ecological Research (KPLTER) site than for the rest of the FIFE site, where cattle grazing was a major confounding effect. The stratified design did not appreciably change the estimated site‐wide means for surface climate parameters but accounted for between 25 and 45% of the sample variance depending on the variable. The design was weakened by undersampling of several strata, by high within‐station variance in soil and vegetation data, and by failure to account for diverse land management practices on private lands surrounding KPLTER. We recommend that future large‐scale climatological studies include the development of a digital terrain data base well in advance of field campaigns and that multitemporal imagery be used to obtain preliminary estimates of spatial and temporal variance in surface biophysical properties. We also recommend that sampling for the most heterogeneous biophysical variables be conducted in the framework of a multistage estimation scheme incorporating remotely sensed data. Although this means that ground‐based estimation of regional fluxes cannot be made independent of aircraft or satellite data, it may well be the only means of obtaining reliable estimates of these variables over large areas.