No description available.

Four small watersheds, Kamwamphula, Luelo, Kanyanga and Mphiri, near Emusa (Lundazi District) in Eastern Province, Zambia, were studied to document transitions in land use over time and to project the impacts of land use and topography on runoff, erosion and sediment delivery. Landuse was delineated from 2007 IKONOS image (one meter pixel), and Landsat imagery was used to depict the historic changes in landuse between the period of 1989 and 2007. The GWLF model was used to predict the impact of the landuses on the hydrology of the area. There had been an increase in clearing of forest area mainly due to the expansion of cropland area. The highest rate of clearing was predicted for the Kamwamphula watershed where the forest cover decreased from 95% to 71% over the 18 year period. The GWLF model was used to predict the impact of the landuse on the hydrology and sediment delivery. In comparison with the limited field data available from the four watersheds, the GWLF model gave poor prediction of streamflow, probably because the hydrology of the area is poorly understood and dambo function in the landscape is not well represented in the model. Highest runoff, erosion and sediment yields came from Luelo watershed which has steeper slopes and, less vegetative cover and poor permeability of soils. The GWLF gave poor prediction of streamflow, probably, because the hydrology of the area is poorly understood. / Master of Science

Two aspects of salinity in arid land were investigated as part of the present dissertation: the first was the potential re-use of industrially generated brine for irrigating landscape plants, and the second was the ecological restoration of saline farmland. The following is a summary of the most important points. With water conservation efforts accelerating in arid environments, industrial wastewater is considered a candidate for re-use. We investigated the use of high EC (electrical conductivity) cooling-tower water to irrigate nine common landscape plants in an urban environment. Each plant (replicated in a block design) was irrigated according to water demand determined by the soil moisture deficit, with one of three water treatments: blowdown water (3.65 dS m⁻¹), well water (0.52 dS m⁻¹) and a 1:1 blend (2.09 dS m⁻¹). Results indicate the salinity of the irrigation water did not have a significant effect (P>0.05) on growth or water use but, soil salinities were higher in basins irrigated with blowdown water compared to those irrigated with well water. The overall feasibility of reusing industrial brines to irrigate urban landscapes is discussed in light of the results. Restoring abandoned arid farmland can be challenging because topographic, geomorphic and hydrologic features have been degraded and cannot support a diverse native plant community. Typical amelioration practices depend upon good quality water to restore the soil’s physiochemical properties, however the long-term availability of any water is rare. A mitigation banking project to return 432 hectares of farmland to an open-space designation involved the collaboration of scientists, landscape architects and engineers to achieve five main goals: water management, erosion control, decreasing soil salinity, and increasing species diversity and vegetation cover. Two strategies evolved in the planning process that work in tandem to achieve these goals: a water management system that redirects storm water and run-off to discrete areas of the site for subsurface storage as plant-available water, and the introduction of a diverse mix of native plants. Field trials tested the strategies and also investigated different soil surface treatments, seeding methods and irrigation regimes against the germination and establishment of a customized native seed mix. Results from vegetation data indicate a combination of soil ripping and imprinting leads to the highest germination and establishment rates and drip irrigation helped establish transplanted seedlings. The project was designed so the longterm outcome does not depend on continual inputs and maintenance.

The Weeks Bay watershed in Baldwin County, Alabama has experienced rapid changes in landuse/ landcover (LULC) from 1990 to 2000. These changes have resulted in increased upland erosion and higher concentrations of suspended sediment within the watershed. For this research project a spatial model was developed to identify potential sources of sediment relevant to LULC and slope. Landsat satellite imagery was classified to assess LULC within the Weeks Bay watershed. The classification includes forested vegetation, herbaceous vegetation (seasonal and persistent), mixed/ transitional vegetation, urban/ built-up areas, sparse/ residual vegetation and water, with an overall accuracy of 78%. Change detections of the classified images yielded substantial increases in urban areas (92.5%). These data were coupled with slope data in a geographic information system and a raster analysis provided a qualitative evaluation of potential sediment sources within the Weeks Bay watershed based on the change in LULC and slopes of the landscape.

Intensive agriculture has led to an increase in production; however this has often coincided with a decline in the provision of other Ecosystem Services (ES). ES affected include those regulated by soil chemical, physical and biological properties such as biodiversity provision and the regulation of nutrient cycling, water quality and rates of greenhouse gas emissions. A growing awareness of the value of nonproduction ES to human health and wellbeing has encouraged the funding of agrienvironment schemes in the UK, through which farmers receive funding to alter management practices to increase the provision of certain ES. One particular management change encouraged through agricultural payments is the creation of species rich grassland (SRG) on former intensively managed (IM) arable or grassland sites. Under these schemes farmers are required to carry out an extensification of management practices by reducing or ceasing fertiliser application, grazing and cultivation, or removing the existing crop or sward and sowing a specified seed mix of desired grassland species. Despite the commitment of substantial sums of money and land to extensification schemes, there has been little research into the extent to which they enhance provision of multiple ES and the potential for the legacy of intensive agriculture to limit ES provision and greatly reduce the value of extensification. This study aimed to: 1) compare soil properties between sites remaining under intensive management and those that had undergone extensification; 2) relate soil properties to; fluxes of the greenhouse gas nitrous oxide (N2O), plant diversity, soil microbial diversity and concentrations of nutrients in leachate from intensively and extensively managed sites in order to determine potential benefits of extensification. Paired field plots were established on working farms in south east Scotland and at Rothamsted Research North Wyke in south west England. Each of the four plot pairs in Scotland consisted of a newly created SRG on former arable land and an adjacent IM plot. The SRG plots ranged in age from 3 to nine years old in 2010. Soil samples were collected from the Scottish sites twice yearly in 2010 and 2011, alongside regular measurements of N2O fluxes from soil and assessment of plant diversity. At North Wyke four replicated SRG plots, forming part of an existing experiment on former intensive grassland, were each paired with an IM plot. Soil samples were analysed for their chemical and physical properties and for the concentration of certain phospholipid derived fatty acids (PLFA) biomarkers to compare the composition and size of the soil microbial community. Soil N2O fluxes and the nitrogen (N) and phosphorus (P) concentrations of soil water samples measured in 2010 and 2011. Results from more intensive N2O flux measurements, conducted in 2012, were compared to model output from the UK-DNDC model to assess its potential for predicting changes in N2O emissions following extensification. No significant difference was found in any soil chemical or physical properties between paired IM and SRG plots in Scotland, although soil bulk density tended to be lower in the older SRG plots relative to the paired IM plots. Nitrous oxide emissions were low from all plots with only an occasional emissions peak being recorded and overall there was no significant effect of management intensity on soil N2O fluxes. The UK-DNDC model outputs were generally of a similar order of magnitude but poorly correlated with measured N2O fluxes and soil water and available N content. Botanical diversity was enhanced in the SRGs compared to the IM plots, though plant species were mostly of low conservation value and indicative of a high nutrient environment and the diversity of the SRG plots was low, compared to long-established semi-natural grassland elsewhere in Europe. Total soil PLFA concentration was significantly higher in the IM plots but the fungal concentration and the ratio of Gram positive:Gram negative bacteria were no greater in the SRG, suggesting it had begun to resemble long-term unimproved grassland. Despite limited success at obtaining soil water samples, at North Wyke concentrations of mineral N in soil water were lower from the SRG plots than the IM plots, although there were no consistent differences in total P or organic N concentrations, organic N contributed over 80% of the total N in soil water samples from the SRG plots. This study has shown that the legacy of intensive agriculture continues to affect soil properties for at least 10 years following extensification. The results suggest that the potential for newly created SRGs to provide enhanced ES’ could be limited and may not justify the reduction in productivity and the financial input associated with shortterm extensification schemes.

No description available.

To assess the impacts of non-point source pollution on Ontarios wetlands I examined the landuse in the watershed, the water and sediment quality, and the aquatic vegetation in 22 wetlands. I characterized the primary contaminants that enter marshes in Ontario and relate their concentrations to the relative amounts and types of land use in their watershed. I measured levels of nutrients, suspended particulates and trace organics in water and sediment. Species richness and structural diversity of the vegetative community was used as an indicator of wetland health. I included a comparison of two techniques, immunoassays and gas-chromatography (GC) to measure trace organics (PAHs and metolachlor; Chapter 1). Comparison of results from both techniques indicated that immunoassays overestimated analyte concentration by approximately a third. I utilized the immunoassay results, along with selected water quality variables, to characterize contaminants entering these wetlands (Chapter 2). Watersheds containing greater than 95% agriculture contributed the highest suspended particulate, compared with those dominated by urban or forested land. Using multivariate statistics I identified important water and sediment variables that structured these wetlands to be: total phosphorus (TP), ammonia nitrogen (TAN), suspended solids (TSS), specific conductance (COND), sediment phosphorus CTPsed) and inorganic material (lnorgsed) in the sediment. Of these TP, COND and Inorgsed were negatively related to the submergent plant community, while TPsed was positively related. There was no relationship between water quality and the floating or emergent plants but the structural diversity of the plant community was negatively affected by poor water quality. The negative impact of developed land in the watershed on the health of the aquatic plant communities underscores the need to maintain natural areas to trap nutrients and sediments in runoff. / Thesis / Master of Science (MS)

No description available.

Doctor of Philosophy / Department of Geography / Melinda D. Daniels / This dissertation increases our understanding of the drivers that shape and maintain grassland streams and their watersheds by examining the influence of grazing management practices on suspended sediment concentrations, bare ground production, and changes to channel geomorphology. Chapter 2 demonstrates that cattle grazing produces significantly higher baseflow suspended sediment concentrations relative to bison grazing. Suspended sediment concentrations within bison-grazed streams are similar to ungrazed streams, indicating that the substitution of cattle for bison has resulted in degradation of baseflow water quality in grassland streams. Burning frequency, discharge, and seasonality are also significant drivers of suspended sediment concentrations, but are generally less influential than grazing treatments. Chapter 3 indicates that high density cattle grazing treatments produce more bare ground within the riparian zones of grassland stream networks, particularly underneath tree canopy cover. The increased bare ground coverage within riparian areas is correlated with increased suspended sediment concentrations during baseflow conditions, while watershed-scale bare ground production is correlated with increased suspended sediment concentrations during storm flow events. Chapter 4 demonstrates channel geometry and sedimentology are significantly influenced by grazing treatments. This dissertation is the first study to comparatively evaluate the relative influence between cattle and bison grazing on stream geomorphology within any environment. Insight gained from this project can be used by public and private land use managers to improve the environmental integrity of native grassland ecosystems.

This research aims to understand the natural and human influences on erosion in three tributary watersheds to the Mekong River, Yunnan Province, China and to assess the utility of a novel application of isotopic indicators of erosion. It explores how erosion varies through time and space as a function of physical characteristics of the landscape, tectonic forces, and human alteration of the landscape for forestry and agriculture. To accomplish these goals, I use four sediment-associated radionuclides: in situ 10Be, meteoric 10Be, 210Pbex, and 137Cs. These isotopes accumulate in or on sediment grains, and each accumulates to a different depth on the landscape and has a different half-life. Thus, the isotopes can be used to track sediment as it moves across Earth's surface, each providing unique insight into processes occurring over a certain time period (from ~50 to 50,000 years) or eroding to a certain depth on the landscape. The studied watersheds range from 22° to 27° N latitude, and from 200 to 2500 km2 in area. I collected 54 samples of river-borne sediment within the three study watersheds, and measured the concentration of each isotope in every sample. In addition to the measured isotopic concentrations, I utilize on over 20 years of daily sediment yield data at the outlet of each watershed, hillslope steepness, normalized channel steepness (ksn), contemporary land-use data, elevation, and 56 years of mean annual precipitation data (MAP). Long-term erosion rates scale with topographic parameters in two of the three study basins, indicating that topography, or the underlying tectonic forces responsible for topography, control erosion rates over the past 6,000 to 50,000 years. Isotopic data also show that contemporary erosion is higher in cultivated areas than un-cultivated areas, a direct result of agricultural practices. Contemporary sediment yield, however, has not increased notably due to land-use change; however, under-representation of large stochastic events and sediment trapped by agriculture have reduced sediment yield relative to the long-term average in two of the studied watersheds. Overall, the data imply changes in contemporary erosion that are consistent with Chinese policies that promoted deforestation from the 1950's to the late 1980's and conservation from the late 1990's to present. This proves to be a significant finding, as the result of the top-down approach China has taken with conservation policy has been widely called into question in previous studies. While each isotope has the potential to provide unique information regarding erosional processes, in situ 10Be and 210Pbex proved to be the most useful, while meteoric 10Be was the most challenging to utilize. Though interpretation is complex, measuring all four isotopes on the same sediment samples helps to fully realize the potential of in situ 10Be to estimate background erosion by simultaneously allowing for assessment of contemporary and human induced erosion.