Biological and physical data from river sites across a gradient of catchment area under grazing in northern Tasmania

Data Creation
Site Selection: Potential stream sites in northern Tasmania, Australia, were screened by catchment size, northing and slope, and according to attributes aimed at minimising confounding variables, maintaining broad consistency in landscape and geomorphological context, and promoting independence among sites (see Magierowski et al 2015 for more detail). A set of survey sites was selected across a gradient from low to high proportion of land under grazing in their upstream catchments. The extent of grazing land-use was determined from the Bureau of Rural Sciences land-use data layer held by the Department of Primary Industries, Parks, Water and Environment and data on public production and plantation forests supplied by Forestry Tasmania. Land-use categories were based on amalgamations of the Australian Land-use and Management (ALUM) classification. The total area (km2) of each land-use type in each catchment was calculated using ArcMap V.10 and converted to a percentage of the total catchment area. Twenty-seven sites were deemed suitable for the survey, with grazing representing 0 to 80% of the land-use in their upstream catchments; the areas of upstream catchments ranged from 20 to 120 km2 (stream order ranged from 4 to 6). Biological and physical data were sampled in riffle habitat at these 27 sites over the summers of 2008/2009 and 2009/2010. We did not detect any effect of year in any of our analyses, so data were pooled across years.
Biological data selection: Macroinvertebrates were sampled using a 0.1 m2, 250 x 956 mm Surber sampler, and 15 samples were taken at random positions within the riffle zone at each site. Samples were fixed in 5% formalin before laboratory processing. All macroinvertebrates within a 20% sub-sample of the organic matter were identified to family and counted, with individuals from the insect orders Ephemeroptera, Plecoptera and Trichoptera identified to genus/species. Proportion of algal cover was visually estimated to the nearest 0.25% from 15 replicate samples arbitrarily located within the riffle zone using a 0.1 m2 quadrat demarcated by wires into a 10 × 10 grid. In addition, 15 arbitrarily located submerged cobbles in the riffle zone were selected for benthic chlorophyll a sampling. The upper surfaces of these cobbles were scoured using 45 mm2 surface area scourer pads, and the pads kept frozen and in the dark until processing. Chlorophyll a concentration (mg/m2) was estimated by pooling the 15 samples and extracting pigments using 90% acetone; absorbance was measured at 663 nm for chlorophyll a and at 750 nm to remove the effect of turbidity.
Physical data selection: A set of physicochemical variables that had been identified in a separate data-mining exercise as potential land-use drivers of change in Tasmanian stream benthic macroinvertebrates were sampled at each gradient survey site. Water temperature and conductivity were measured at the time of sampling, and under base flow conditions, with a WTW 315i conductivity meter and Tetracon 325 probe. Turbidity was measured with a Hach 2100P Turbidimeter, and water samples were collected for laboratory analyses of pH, total alkalinity, nitrate+nitrite (NOx), dissolved reactive phosphorus (DRP), total nitrogen (TN) and total phosphorus (TP; Analytical Services Tasmania). Overhead shading was measured with a hemispherical densiometer as proportional canopy cover, and the proportion of fine sediments within the sampled riffle zone was estimated using the AusRivAS rapid assessment protocol for substratum composition (percent cover of sand (0.06 – 2 mm) and silt (< 0.06 mm). Two variables representing flow modification were sourced from Tasmania’s Conservation of Freshwater Ecosystem Values (CFEV) database, in turn derived from data held in the Water Information System of Tasmania (WIST) and by Hydro Tasmania. The “accumulated abstraction index” was the sum of all upstream abstractions and diversions divided by the long-term (20 to 35 year) mean annual runoff, and the accumulated regulation index was the sum of all upstream catchment storages (e.g. farm dams) divided by the long-term mean annual runoff.