Organic carbon stores and processes in an upland Paddys River catchment (ACT, Australia) : towards a holistic approach to the assessment of catchment condition

Abstract

It has been recognized that a whole-catchment perspective is required for the assessment of the river condition and river management. Organic carbon (OC) is a life-limiting resource and its flow and processing represent a crucial link between terrestrial and aquatic environments of the catchment, making OC a candidate indicator for the assessment of catchment condition. Human activities tend to lead to substantial changes in catchment vegetation (i.e. the source of OC for the catchment) which might affect OC stores and processing along the OC flow-path. In this thesis, I have synthesized and examined a catchment-based approach to ecological assessment. This approach is based on OC and it follows the flow of OC in an upland catchment – from vegetation to soil and through soil to the river. By assessing the relative importance of land-use and abiotic factors as controls of OC stores and processing I aimed to determine whether major changes in vegetation, following land-use conversion, have affected the stores, retention and recycling of OC in soils of an upland catchment and in an upland river. Considering time and other resources available for the project, the scope of the thesis was limited to four components of the movement of OC in the catchment of the upland Paddys River (ACT, Australia). The ‘soil studies’ of the thesis are concerned with the effect of land-use (conservation, grazing, pine-plantation) on 1) OC stores in soil and 2) OC leaching from soil. The ‘river studies’ of the thesis are concerned with the effect of land-use on 1) reservoirs of OC and 2) OC processing in the Paddys River. To differentiate between the effect of land-use along the Paddys River and the natural longitudinal change in abiotic and biotic river characteristics, the longitudinal patterns in OC reservoirs and processing in the Paddys River are compared with longitudinal patterns in OC reservoirs and processing in four reference rivers. The concentration of OC in soil represents a direct link between catchment vegetation and soil. In Chapter 2, I compared soil OC concentrations across the three types of land-use present in the Paddys River catchment and examined the relative importance of land-use, vegetation structure and abiotic site characteristics (slope inclination and slope aspect, elevation, clay and silt content) in determining soil OC concentration. The presence of charcoal in samples, discovered during sample preparation, provided an opportunity to examine whether missing the presence of charcoal would have led to a different conclusion regarding the dominant factor controlling the concentration of soil OC in the catchment. This question has wide implications considering the proportion of land affected by fires worldwide and the widespread use of soil OC as an indicator of soil condition. My results have shown that land-use and the proportion of clay and silt were the most influential predictors of soil OC concentration. The presence of charcoal masked the effect of land-use and led to an underestimation of the relative importance of land-use in determining soil OC concentration. OC leaching from soil is a phenomenon associated with the movement of water through the catchment, linking catchment vegetation (primary source of OC), soil (stocks of OC potentially available for leaching) and the aquatic environment (potential recipient of terrestrial OC). Using soil cores and a desktop rainfall simulator, in Chapter 3 I assessed the relative importance of land-use and soil abiotic and biotic properties (pH, bulk density, texture, moisture, soil OC, microbial biomass and microbial community composition) in determining the concentration of leached OC and the relative proportion of soil OC leached during a high rainfall event. I found that land-use affected the concentration of leached OC directly and indirectly, through its effect on soil OC stocks which was the most important predictor of leached OC concentration. Land-use also affected microbial biomass and microbial community composition which in turn appeared to be associated with differences in leached OC. Soil texture had a low direct effect on OC leaching; however texture might also affect OC leaching indirectly, through its effect on soil OC stocks, as found in Chapter 2. Bulk density was the most influential physical soil property and a predictor of both OC leaching and microbial biomass. Changes in catchment and riparian vegetation might have an effect on OC inputs, stores and processing in rivers, which underpin river functioning. In Chapter 4, I examined longitudinal patterns in OC reservoirs (total organic carbon (TOC)), suspended OC, drifting coarse particulate OC (drift OC), organic debris and in-stream primary biomass in the Paddys River and four reference rivers during base-flow conditions. The studied rivers represented a range of geomorphologies and a gradient of land-use conversion. My predictions regarding longitudinal patterns in rivers were not confirmed. Rivers draining catchments with conservation land-use did not show consistent patterns in OC reservoirs that would differ from rivers draining a combination of conservation and grazing land. ‘River’ was a more influential predictor of OC reservoirs than the proportion of riparian canopy cover, with TOC being the main determinant of groups in the multidimensional scaling plot. My results suggest strong within-catchment controls of TOC concentration during base-flow conditions which might be maintained despite catchment impairment. These findings have implications for river rehabilitation, when using a reference river as a ‘guiding image’. Decomposition of organic matter has been proposed to be used as a functional indicator of stream condition. It has also been suggested to use the cotton-strip (CS) assay as a standardized method of measuring decomposition rates. In Chapter 5, I examined temporal and spatial variability in CS decomposition rates and the effect of land-use on CS decomposition along the Paddys River and four reference rivers. There was a consistent longitudinal pattern in CS decomposition rates throughout the year in the Paddys River with lowest rates found in headwaters and at the river mouth and highest rates in the middle reaches. My result suggest that diel temperature range, known to be affected by riparian clearing, was the main driver of the observed longitudinal pattern, except for the upper section, where high sediment load limited CS decomposition rates. The proportion of land converted to pasture was a significant predictor of CS decomposition rates, however, its effect had an unclear trajectory. More importantly, the results of this study have suggested that there are further problems that need to be addressed, if the CS assay is to be used as a standardized assessment technique. These include: 1) not well understood relationship between time (i.e. the duration of cotton-strip deployment) and CS decomposition rates; 2) not well understood relationship between temperature and diel temperature oscillation and CS decomposition rates; and 3) large within-site variability in CS decomposition rates which might prevent comparisons with reference conditions. The importance of temperature range also has implications for stream rehabilitation, showing that restoring the temperature regime contributes to the restoration of a fundamental ecosystem process.

Date of Award	2013
Original language	English
Supervisor	Fiona DYER (Supervisor) & Bill Maher (Supervisor)