Andean sediment response to natural and anthropogenic changes at the catchment scale

Abstract

Changing climate and population growth poses a challenge for the water, food and energy security nexus, particularly in glacier-fed mountainous catchments. Soil erosion, driven by both natural and anthropogenic activities, further exacerbates the challenge for resource security. Soil erosion is contributing to increased levels of sediment in river systems threatening crucial river functions, including crop irrigation, drinking water, and hydroelectricity. Past studies in the Andes regarding soil erosion have indicated that human activities have accelerated soil erosion and rates often exceed natural soil production. To date, the majority of Andean studies have primarily focused on larger eastern catchments, with fewer studies focusing on the western catchments. Although, western Andean catchments are often smaller, many major Andean cities are reliant on upon their upstream water supplies. The identification of sediment sources and improved understandings of in-channel sediment is crucial in the management of water security in Andean mountainous catchments. This thesis presents a new land cover map for the Rio Santa catchment, Peru, that builds on freely available spatial data and includes more specific land cover types than past maps. The Rio Santa catchment, Peru, is a catchment with various land cover types ranging from agriculture, forest and glaciers, all with the potential to threaten food, water and energy security if not managed appropriately. High altitude mountainous catchments often pose a challenge to the accessibility of sampling sites to record ground truth spatial data. However, by combining ground truth points and points created in Google Earth a variety of land cover types could be recorded and spatially mapped. The use of freely available software to record land cover types also promotes repeatability and provides a method for recording land cover data in other mountainous catchments. This thesis aimed to assess the natural and anthropogenic factors contributing to in-channel sediment at the catchment and sub-catchment scale. A distributed sampling approach was used along the Rio Santa catchment and the Ranrahirca glaciated sub-catchment to investigate the contributors to sediment production for this Andean system. This distributed approach permits the quantification of key sources throughout a catchment at both scales. The key sediment sources identified and investigated included glacial sediment, agricultural land, forestry operations, and land under natural vegetation. All source and mixture samples were analysed using Wavelength Dispersive X-ray Fluorescence (WD XRF) to produce geochemical fingerprints. Following this, the geochemical fingerprints were used alongside the MixSIAR unmixing model to apportion sediment sources for both the Rio Santa catchment and Ranrahirca sub-catchment. For the Rio Santa catchment, the greatest contributor to sediment is the lake at the source of the catchment, followed by sediments produced in the Cordillera Negra. This suggests that sediments from glaciated sub-catchments are less important for the Rio Santa than sediment from non-glaciated areas. In the case of the Ranrahirca sub-catchment, which for this investigation was divided into upstream of the national park boundary and downstream of the boundary due to the pattern of land cover, the greatest contributor to upstream sediment production is channel bank. Furthermore, upstream sediment sources contributes more to sediment in the downstream portion, followed by local arable land and channel bank. This suggests the glaciated upstream greatly influences sediment throughout the Ranrahirca sub-catchment. Sediment sampling in high altitude and remote catchments such as the Rio Santa poses many challenges, but sediment fingerprinting can provide insight into catchment-wide sediment, providing representative source samples are collected. Furthermore, sediment fingerprinting approaches, when incorporated into to local resource management decisions, have the potential to reduce the degradation of catchments by providing an insight into catchment sediments.