Stream monitoring using near-infrared spectroscopy of epilithic material

The European Union Water Framework Directive (WFD), with initiatives to manage surface water resources, has increased the need for fast and inexpensive methods for monitoring conditions in streams. The objective of this thesis is to assess the potential of near-infrared spectroscopy (NIRS) of epilithic material to become such method. NIRS, which is a technique that is commonly used in industry for process monitoring and quality control of products, registers the chemical properties of organic material on a molecular level. Epilithic material, i.e. the layer of dead and alive material that covers stone surfaces in streams, is continuously influenced by the stream water that flows over it, and it has the potential to integrate chemical and biological conditions over time. The temporal integration is a significant factor, since conditions in streams can change within hours or days. The thesis consists of two published papers. In the first paper a new sampler for epilithic material, the Stone Brusher, was described and the performance evaluated. The Stone Brusher is designed to take qualitative or semi-quantitative samples of epilithic material from stones at 7–50 cm water depth. The epilithic material is dislodged from the stone surface with a rotating brush enclosed in a chamber, and the material is drawn up directly into the sample bottle with an air-cylinder. The operator takes a sample quickly and without putting hands into the water. The sampler is made of plastic, stainless steel and aluminium and weighs 3.1 kg. It is designed to meet the demand for standardized sampling for research and environmental monitoring and to improve working conditions for sampling personnel. The equipment allows sampling from surfaces of bedrock and large stones that cannot be lifted from the bottom. Using data of near-infrared spectroscopy and diatom analyses, this new sampler was evaluated in comparison to the toothbrush method, a primitive method which is the current standard in EU. The results indicate that the Stone Brusher reduces sampling variability compared with the toothbrush method. In the second paper, the Stone Brusher was used to collect epilithic material from 65 sites (42 uncontaminated and 23 contaminated) from streams in the widespread mining area called the Skellefte-district in Västerbotten, northern Sweden, in order to test the hypothesis that impact on the epilithic material caused by emissions from mining and mining-related industries can be detected using NIRS. The epilithic material was filtered onto glass fibre filters, measured by NIRS, and the results were modelled using Principal Component Analysis (PCA). The NIRS approach was evaluated by comparing it with the results of chemical and diatom analyses of the same samples. Based on PCA, the NIRS data distinguished contaminated from uncontaminated sites and performed slightly better than chemical analyses and clearly better than diatom analyses. Of the streams designated a priori as contaminated, 74 % were identified as contaminated by NIRS, 65 % by chemical analysis and 26 % by diatom analysis. Unlike chemical analyses of water or of epilithic material samples, NIRS data reflect biological impacts in the streams. Given that, and the simplicity of NIRS-analyses, further studies to assess the use of NIRS of epilithic material are justified. NIRS has the potential to become a fast method for screening in regions where large numbers of streams occur to find impacted streams or as a routine method for temporal monitoring in selected streams for early detection of environmental impact, similar to process monitoring in industry