Processes controlling soil hydrology and pedogenic carbonate formation in loess tablelands, Nebraska, USA

Abstract

As part of a larger project, the objective is to understand how microtopography and the presence of buried soils potentially influence the stability of the landscape in semiarid southwestern Nebraska. It was hypothesized that microtopographic features such as raised rims, depressions, and troughs found on the summits of loess tablelands direct water towards the center of the table and prevent further erosion by the gullies. The presence of a loess-paleosol sequence could provide additional landscape stability because the finer-textured paleosols could have a higher water holding capacity than the surrounding loess, which could increase vegetation cover and thus, decrease wind and water erosion. To better understand the varying degrees to which the microtopographic features and the presence of the paleosols impact landscape stability, soil properties along a three-point transect were analyzed. The three-point transect begins at the highest elevation on the edge of a table closest point to loess source (PT1), and it ends inward at the lowest elevation towards the center of the table in a trough, which is the farthest point from the loess source (PT3). Results from these soil analyses showed differences in loess sedimentation and pedogenesis across the transect. PT3 had slightly thinner loess deposits, a finer particle size distribution, more visible carbonate and higher total carbon and nitrogen than the other two points on the transect. The finer particle size distribution is likely a result of loess sedimentation because PT3 is farther downwind from the loess source. The finer particle sizes and run-on likely increase vegetation cover and pedogenesis at PT3, while coarser particle sizes and loss of water through surface run-off likely decrease vegetation cover and pedogenesis at PT1. The Brady Soil, a prominent 1-m thick paleosol found up to ~3 m deep at the study site, had a finer particle size distribution compared to other depths at all three points. The upper paleosols closer to the modern surface, did not show distinctly different particle size distributions compared to the loess, so the effect of the paleosols on water retention and infiltration could not be determined from the analyzed soil properties. To better understand the effect of the paleosols, the numerical model, Hydrus 1D, was used to simulate water flow and root uptake through a 4.5 m loess-paleosol sequence (the middle transect point; PT2) using measured soil hydraulic parameters. In addition to the loess-paleosol sequence at PT2, water flow was also modeled through a hypothetical profile in which all paleosols were removed. Water flow was simulated through the two profiles for four different meteorological scenarios: two wetter years, a drier year and an average year. While the measured soil hydraulic properties confirmed that the paleosols had distinctly different properties than the surrounding loess, model results indicate the primary control on volumetric water contents (VWC) at depths < 2 m was meteorological conditions. In both modeled profiles, the dry year had lower VWC and less root uptake compared to the other three other years. The removal of the paleosols resulted in only a minor decrease in root uptake and as a result, the paleosols likely do not have a large influence on geomorphic stability. The model results also show that infiltration rarely reaches depths > 2 m in the profile, indicating that the prominent Brady Soil is minimally affected by wetting in the modern climate. Although the paleosols have minimal impacts on root uptake and tableland stability, the model results did confirm that layering of the distinctly different hydraulic properties altered water redistribution by slowing infiltration. The slowed redistribution likely increases the groundwater recharge lag time and influences the depth at which pedogenic calcium carbonates form (carbonate). The impact of soil hydrology on depth of carbonates was investigated further by coupling the previously mentioned model with a major ion chemistry model. The simulated depth of carbonate accumulation was ~90, within the second paleosol, confirming that carbonates found deeper in the soil profile likely formed in past climates. These results were combined with microscopy analyses to further evaluate factors complicating use of carbonates as a paleoclimatic proxy. The carbonate morphology observed in the thin sections was mostly needle fiber calcite suggesting that the physiological processes of fungi played a large role in carbonate formation and accumulation. These results indicate that both soil hydrology and biological processes need to be considered before carbonates can be used as a paleoclimate proxy.