Provenance & quality
Originally designed to improve understanding of water use by coniferous forests, monitoring within the Plynlimon/Vyrnwy research catchments has subsequently developed into a multi-disciplinary project underpinning a wealth of hydrological research. The stream hydrochemistry dataset is a combination of chemical determinands and in-situ physical properties. Samples are analysed in laboratories for approximately 50 chemical determinands. Rainfall was sampled automatically using a sequential, continuously open collector mounted above an enclosed carousel of bottles sufficient to collect all the rain falling over each seven-hour period occurring in a week. Instantaneous samples of stream water was also collected every 7 hours using automatic samplers, each fitted with a carousel of bottles sufficient for a week. Each week the samples were returned to the laboratory, filtered, then two subsamples were taken, one was acidified with nitric acid and the other left unacidified. Chemical analysis was by ICP-OES (for major cations, B, S and Si),ICP-MS (for trace elements), ion chromatography (for major and minor anions), potentiometry (for pH), Gran titration (for alkalinity),conductivity meter (for conductivity), Shimadzu analysers (for DOC and TDN) and Seal discrete analyser (for NH4). The complete list of analytes includes F, Br, I, Cl, SO4, NO2, NO3, NH4, TDN, DOC, K, Mg,Na, Ca, B, S, Si, pH, Gran alkalinity, conductivity, Li, Be, Al, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Se, Rb, Sr, Mo, Cd, Sn, Sb, Cs, Ba, W, Pb, U, La, Ce and Pr. The analytical methodologies, detection limits are provided in the accompanying metadata.The Raw Data Archive (denoted by "RDA" at the end of each variable name) is an archive of the raw measurements as reported by the analytical laboratories. These data have not been altered in any way, except to correct glitches that can be traced back to the original laboratory files and analytical sheets and corrected on that basis. We are providing this raw data archive for completeness of documentation, and also in the interests of full transparency concerning data editing that has been subsequently applied. The raw data archive also provides the opportunity to go back to the original data in case new information emerges suggesting that interesting realworld phenomena may have been inadvertently edited out. However, we do not recommend that the raw data archive be used for routine analyses, because the raw data contain some values that are believed to be substantially in error. Instead, we recommend that routine analyses use the edited version of the data, where to the greatest extent possible, problematic data values have been corrected or excluded.The differences between the edited data and the raw data archive are explained in detail in the accompanying metadata file, "Plynlimon 7hour data editing documentation.pdf"