Assessment of the effects of upstream land uses and riparian vegetation composition on surface water quality of lowland streams : A thesis submitted in partial fulfilment of the requirements for the Degree of Master of Water Resource Management at Lincoln University

Non-point sources pollution caused by land-based developments (such as increased in agricultural land, residential and industrial areas) become the major threats to the freshwater quality around the world. In New Zealand, the surface water quality has also been declining, and a significant increase in sediment and nutrients are considered as major water quality problems. Riparian plantings (vegetation along the riverbank) are recommended as a cost-effective measure because they could reduce sediment and nutrients (nitrogen and phosphorus) inputs from non-point sources through their main functions: infiltration, filtration and absorption. Also, restoring and managing riparian plantings along the waterways is being introduced as one of the best management practice in New Zealand to reduce the impacts of catchment land use on water quality. However, the effectiveness of riparian plantings may vary in accordance with vegetation compositions (such as shaded buffer and grassland buffer) and the width of riparian planting area. This study aims to assess the relationship between upstream and sub-catchment land-use and water quality and the effectiveness of different riparian vegetation compositions (shaded, un-shaded/grassland and unplanted areas) in reducing nutrient and sediment inputs from upstream and sub-catchment contributing land uses in lowland streams. The Styx River catchment in Christchurch, which has a wide range of riparian vegetation compositions, was chosen as study area. A total nine sampling sites were selected based on three different riparian vegetation compositions (shaded, grassland and unplanted) at the Styx and its main tributaries: Smacks and Kaputone Creeks. A total of 72 water samples were collected over eight dates (fortnightly over five dates and three dates after rain events) from nine sampling sites. Sub-catchment land uses were determined by using Arc GIS software, and defined into four main types: cropland, pastoral land, forested land and built-up area. Riparian plantings (both shaded and grassland) showed a positive effect on reducing the concentrations of conductivity, turbidity, sediment, phosphorus and nitrogen. The riparian plantings with trees showed more effectiveness in reducing pollutants than grassland areas because mostly, the lowest levels of pollutants (conductivity and phosphorus, turbidity, sediment and nitrogen) were found at shaded sites. However, the proportion of sediment and nutrient fluxes depends on discharge rate. Dissolved oxygen levels showed correlation with water temperature levels. The upstream and sub-catchment land-use (especially built-up area and pastoral land) was found to have a positive relationship with conductivity and nitrogen (specifically nitrate, total nitrogen and total dissolved nitrogen). The > 5 m wide riparian areas showed more effectiveness in reducing contaminats from built-up and pastoral influence area than solely pastoral land influence area. In order to determine the effectiveness of riparian plantings, a number of factors (such as length and width of riparian plantings’ area and stream shaded area) need to be considered, and a balance between the main functions of the riparian plantings in relation to the sensitivity of a proposed site will also need to be considered. Furthermore, the research suggests that the cooperation of private land owner is critical in establishing and managing riparian plantings along waterways.