The use of the sediment fingerprinting technique to quantify the different sediment sources entering the Whangapoua Estuary, North Island, in New Zealand
Roddy, B. P. (2010). The use of the sediment fingerprinting technique to quantify the different sediment sources entering the Whangapoua Estuary, North Island, in New Zealand (Thesis, Doctor of Philosophy (PhD)). University of Waikato, Hamilton, New Zealand. Retrieved from https://hdl.handle.net/10289/4485
Permanent Research Commons link: https://hdl.handle.net/10289/4485
New Zealand estuaries are sites of ecological, economic, and recreational significance. Estuaries are vulnerable to the impacts of increased erosion as they act as natural sediment traps. The objectives of this study were to; 1) quantify the relative amounts of sediment entering the estuary from the native forest, exotic forest, and agricultural landscape units in the Whangapoua catchment, Coromandel Peninsula, New Zealand; 2) identify the dominant processes generating the sediment within the native forest, exotic forest, and pastoral landscape units; and 3) assess the utility of the sediment fingerprinting technique in New Zealand by comparing the results with other sediment measurement techniques. Sediment fingerprinting, which uses geochemical elements to link potential source areas to the estuary sediment, was used to identify sediment sources in the Whangapoua catchment. Three landscape units (referred to as native forest, exotic forest, and agriculture), and three erosion positions (surface, subsurface, and streambanks) were investigated. A radionuclide tracing study, a stream suspended sediment monitoring programme, and catchment modelling were undertaken to compare with the sediment fingerprinting results. An initial pilot study was undertaken which confirmed that sediment fingerprinting could distinguish sediment derived from the three landscape units and three erosion positions. In a full sampling programme, the landscape units and erosion positions in the Whangapoua catchment were each characterised by analysing 50 samples using ICP-MS to determine the concentrations of 29 elements. The elements Si, P, Se, V, U, In, and Bi were identified as forming a composite fingerprint to distinguish landscape units. The native forest landscape unit (21% of the catchment area) contributed 62% of estuary sediment, with 23% from the exotic forest (61% of the catchment area), and 15% from the agricultural landscape unit (18% of the catchment area). The elements Se, Fe, Ba, Mn, P, and Ca formed a composite fingerprint to distinguish erosion positions and showed that most of the estuary sediment was derived from subsurface (79%), followed by streambanks (13%), and then surface sources (8%). A radionuclide tracing study was undertaken and 15 samples were used to characterise each of the surface, subsurface, and streambank erosion positions. 137Cs was effective at distinguishing the surface (0-2cm) from other erosion positions, but could not distinguish the subsurface (>20cm) from the streambank sediment. The 137Cs results indicated that up to 98% of the estuary sediment was derived from subsurface and streambank sources. Stream suspended sediment monitoring was undertaken at four small subcatchment sites over a two year period. Monitoring commenced at the four sites as follows; immediately harvested exotic pines, six month post harvested pines, ten year re-growth pines, and agricultural pastures. The erosion rates were high in pines after harvesting (48 t km-2 yr -1), but dropped to 28 t km-2 yr -1 six months post harvesting and then to 2 t km-2 yr -1 ten years post harvesting. The agricultural erosion rate was calculated at 7 t km-2 yr -1. The lack of a native forest monitoring site, lack of replication, data capture problems, and inherent errors limit confidence in the stream suspended sediment monitoring results which should only be considered indicative. A stream bed sampling programme used 18 sampling points to estimate how much fine sediment was stored within the rivers and streams. Less than 4% of the annual estuary fine sediment budget was stored within the stream and river beds, indicating that sediment was efficiently conducted from the native forest landscape unit into the estuary. Two catchment models (Sediment Yield Estimator and the New Zealand Empirical Erosion Model) gave results that were contradictory to the sediment fingerprinting results as they suggested that deforested areas (such as agriculture and harvested pines) were sediment generating areas. This thesis demonstrated that the use of sediment fingerprinting in New Zealand was an effective means of quantifying various sediment sources based on landscape unit and erosion position. The current pattern of landuse within the Whangapoua catchment is appropriate in providing an economic return while minimising the levels of sediment delivery to the Whangapoua estuary.
University of Waikato
All items in Research Commons are provided for private study and research purposes and are protected by copyright with all rights reserved unless otherwise indicated.
- Higher Degree Theses