Search

Terrigenous sediment delivery into tropical coastal waters from unpaved road development is a key stressor influencing the global decline of coral reef ecosystems. Higher turbidity of coastal waters from this terrigenous sediment can result in smothering, bleaching, and reduced photosynthetic activity for coral reefs. Since the 1990’s, the small Puerto Rican municipality of Culebra has experienced significant rates of land development, and an expansion of its unpaved road network, which has been identified as the contributing factor responsible for the continuous decline of the island’s live coral cover. This research focuses on quantifying runoff, and sediment production rates for Culebra’s unpaved road and undisturbed surfaces at the plot (~3 m²), and watershed (~1 km²) scales. A total of 46 rainfall-runoff simulations, and 119 Guelph permeamter tests were conducted to collect this data. Results indicate that only 1.2 mm of rainfall is required to produce overland flow on unpaved roads. At least 7.9 cm of rainfall is needed to generate overland runoff on undisturbed hillslopes. Erosion rates were greatest for recently graded road segments, with an average annual erosion rate of 54.65 Mg ha⁻¹ yr⁻¹. Unpaved roads that have been left ungraded for over one year erode at about half this rate (29 Mg ha⁻¹ yr⁻¹). Undisturbed surface erosion was negligible in comparison, with a mean rate of 0.05 Mg ha⁻¹ yr⁻¹. Based on this data, there are up to five orders of magnitude difference between erosion rates of graded unpaved roads and natural soil surfaces in Culebra. Within one year, a total of 335 storms occurred between. Around a quarter of all recorded storms produced overland runoff on unpaved road surfaces. Hurricane Maria was estimated to be the only storm that year to generate overland flow on undisturbed surfaces. ArcGIS was used to identify segments of road most likely to transport overland runoff to the coastline, as well as segments generating the most erosion. Sediment production rates for the study areas ranged between 3.1 and 10.9 Mg yr⁻¹. This information is beneficial towards documenting the overall effectiveness of erosion mitigation structures built to reduce terrigenous sediment transport produced from these roads

Terrigenous sediment delivery into tropical coastal waters from unpaved road development is a key stressor influencing the global decline of coral reef ecosystems. Higher turbidity of coastal waters from this terrigenous sediment can result in smothering, bleaching, and reduced photosynthetic activity for coral reefs. Since the 1990’s, the small Puerto Rican municipality of Culebra has experienced significant rates of land development, and an expansion of its unpaved road network, which has been identified as the contributing factor responsible for the continuous decline of the island’s live coral cover. This research focuses on quantifying runoff, and sediment production rates for Culebra’s unpaved road and undisturbed surfaces at the plot (~3 m²), and watershed (~1 km²) scales. A total of 46 rainfall-runoff simulations, and 119 Guelph permeamter tests were conducted to collect this data. Results indicate that only 1.2 mm of rainfall is required to produce overland flow on unpaved roads. At least 7.9 cm of rainfall is needed to generate overland runoff on undisturbed hillslopes. Erosion rates were greatest for recently graded road segments, with an average annual erosion rate of 54.65 Mg ha⁻¹ yr⁻¹. Unpaved roads that have been left ungraded for over one year erode at about half this rate (29 Mg ha⁻¹ yr⁻¹). Undisturbed surface erosion was negligible in comparison, with a mean rate of 0.05 Mg ha⁻¹ yr⁻¹. Based on this data, there are up to five orders of magnitude difference between erosion rates of graded unpaved roads and natural soil surfaces in Culebra. Within one year, a total of 335 storms occurred between. Around a quarter of all recorded storms produced overland runoff on unpaved road surfaces. Hurricane Maria was estimated to be the only storm that year to generate overland flow on undisturbed surfaces. ArcGIS was used to identify segments of road most likely to transport overland runoff to the coastline, as well as segments generating the most erosion. Sediment production rates for the study areas ranged between 3.1 and 10.9 Mg yr⁻¹. This information is beneficial towards documenting the overall effectiveness of erosion mitigation structures built to reduce terrigenous sediment transport produced from these roads

Located in central Brazil, the Araguaia River is the largest river flowing through the Cerrado, the Brazilian savanna. The thesis presented here assesses the hydro-geomorphology of the middle Araguaia River-floodplain system by characterizing flooding patterns and linking these patterns to the geomorphology of the floodplain. It also determines the response of floodplain lake morphometry and surface water connectivity to the annual flooding of the river, and describes how different floodplain geomorphologic units influence changes in open water areas in the floodplain from the dry season to the wet season. Peak discharges along the middle Araguaia River can be reduced downstream despite large increases in drainage area and the contribution of tributary inputs. After analyzing average daily discharge measurements from 1975 to 2007 along an upstream reach and a downstream reach in the middle Araguaia River, four main flooding types are characterized based on the magnitude of the peak discharge and the pattern of peak discharge reduction that occurs as the flood wave moves downstream. Short-term losses of channel discharge during the flooding peak and over the flooding season from November to May are estimated, with the downstream reach displaying more short-term channel loss compared to the upstream study reach. Differences in floodplain geomorphological characteristics between the two study reaches, including the proportions of distinct geomorphologic units (a lower elevation impeded floodplain, a unit dominated by paleomeanders, and a unit of accreted banks and islands), influence the patterns of peak reduction and channel loss. Short-term losses of channel discharge during flooding peaks are usually re-gained by the channel by the end of the flooding season, although in two years about 10% of the volume input into the downstream reach was lost from the channel over the flooding season. Using satellite imagery and an open water index, changes in lake area, perimeter, and surface water connectivity with the main channel between dry season and the wet season are determined for 32 floodplain lakes. The changes in lake morphometry and connectivity are linked to how fluvial processes formed the floodplain lakes. Spatial variations in the floodplain areas that became open water from the dry season to the wet season demonstrate that distinct floodplain geomorphologic units influence the extent and location of open water areas during flooding. Floodplain lakes that expand in area and in depth and are connected to the river channel via surface water likely provide storage areas for the channel losses and peak discharge reductions observed in some of the flooding types for the middle Araguaia River. Although there have been attempts to plan the placement of dams on the Araguaia River, the river is not impounded, allowing for the analysis of a river system with an unaltered flow regime. This thesis contributes to knowledge of a large and understudied tropical river in an ecologically sensitive region.

Located in central Brazil, the Araguaia River is the largest river flowing through the Cerrado, the Brazilian savanna. The thesis presented here assesses the hydro-geomorphology of the middle Araguaia River-floodplain system by characterizing flooding patterns and linking these patterns to the geomorphology of the floodplain. It also determines the response of floodplain lake morphometry and surface water connectivity to the annual flooding of the river, and describes how different floodplain geomorphologic units influence changes in open water areas in the floodplain from the dry season to the wet season. Peak discharges along the middle Araguaia River can be reduced downstream despite large increases in drainage area and the contribution of tributary inputs. After analyzing average daily discharge measurements from 1975 to 2007 along an upstream reach and a downstream reach in the middle Araguaia River, four main flooding types are characterized based on the magnitude of the peak discharge and the pattern of peak discharge reduction that occurs as the flood wave moves downstream. Short-term losses of channel discharge during the flooding peak and over the flooding season from November to May are estimated, with the downstream reach displaying more short-term channel loss compared to the upstream study reach. Differences in floodplain geomorphological characteristics between the two study reaches, including the proportions of distinct geomorphologic units (a lower elevation impeded floodplain, a unit dominated by paleomeanders, and a unit of accreted banks and islands), influence the patterns of peak reduction and channel loss. Short-term losses of channel discharge during flooding peaks are usually re-gained by the channel by the end of the flooding season, although in two years about 10% of the volume input into the downstream reach was lost from the channel over the flooding season. Using satellite imagery and an open water index, changes in lake area, perimeter, and surface water connectivity with the main channel between dry season and the wet season are determined for 32 floodplain lakes. The changes in lake morphometry and connectivity are linked to how fluvial processes formed the floodplain lakes. Spatial variations in the floodplain areas that became open water from the dry season to the wet season demonstrate that distinct floodplain geomorphologic units influence the extent and location of open water areas during flooding. Floodplain lakes that expand in area and in depth and are connected to the river channel via surface water likely provide storage areas for the channel losses and peak discharge reductions observed in some of the flooding types for the middle Araguaia River. Although there have been attempts to plan the placement of dams on the Araguaia River, the river is not impounded, allowing for the analysis of a river system with an unaltered flow regime. This thesis contributes to knowledge of a large and understudied tropical river in an ecologically sensitive region.