Continuous data is collected at 9 sites throughout New Hampshire. At each site data is collected every 15 minutes by the datalogger from the SUNA and YSI. Data is collected and transmitted to UNH by cell telemetry once a day where it is stored on the EPSCoR data server. The data that is collected from the SUNA are Nitrate in M, Nitrate in mg/L (and a host of diagnostic variables such as measurement error, lamp temperature etc.) Data that is collected by the YSI are Cable Power, Stream Temperature, Conductivity, Specific Conductance, pH, Depth, Optical DO in mg/L, Optical DO in %, Turbidity Raw, Turbidity in FNU, FDOM in RFU, and FDOM in QSU. The datalogger also records panel temperature and battery voltage, as well as date and time.
The data processing/quality control procedure is as follows. Sensor data are prone to signal interference due to a variety of factors (for example increased turbidity levels are known to influence estimate of nitrate concentration). As a result, quality assurance/quality control (QA/QC) procedures are required to provide an initial assessment of the reliability and scientific quality of the data collected. QA/QC is performed using a combination of grab sample analysis and data analysis procedures. Water quality grab samples are collected weekly by lab personnel and during storms by ISCO auto-samplers to verify sensor performance. They are essential to evaluate the accuracy of in situ water quality measurements and in the estimation of solute fluxes. The water samples are analyzed at the University of New Hampshire Water Quality Analysis Laboratory.
Quality control (QC) for environmental sensor data begins with preventative measures taken in the field and continues throughout the data processing workflow. The team is incorporating an automated data quality review and flagging procedure to evaluate and flag raw sensor data, generating a data product for provisional release in near-real time (level 1a). Subsequent quality control includes a more detailed data review/flagging process by the data management team (level 1b). Near realtime provisional data displayed in the EPSCoR query tool will be either level 1a or 1b. Higher level data products (level 2 and above; see below), will be available in the future.
Implementation of QC varies for each sensor. A set of rules for automated QC have been developed for each water quality parameter, and flagging may be based on one or more of the following tests: voltage limits, sensor limits, missing values, persistent values, slope exceedance, spatial and internal inconsistency etc, as described in Campbell et al. 2013. All QC tests and data processing history are documented in the associated metadata files.
As an example, data retrieved from the nitrate analyzer are initially checked for integrity based on realistic values from past studies, the optimal operating voltage, sensor operating limits, and potential interference from high turbidity and dissolved organic carbon (DOC) levels and sensor temperature. Water quality grab samples are collected weekly by lab personnel and during storms by ISCO auto-samplers to verify sensor performance. They are essential to evaluate the accuracy of in situ water quality measurements and in the estimation of solute fluxes. The water samples are analyzed at the University of New Hampshire Water Quality Analysis Laboratory. These lab data provide the information necessary to correct for any offsets and/or drifts in sensor data, and are incorporated into level 2 data products.
NH EPSCoR-funded infrastructure of web-servers and availability of custom implemented and publicly available/open-source tools (e.g GCE Toolbox for Matlab) are used to process the data and visualize Level 1 data in near-realtime.
Assessing chemical response in rivers to storm flow in mixed land uses, assessing biogeochemical cycling in river networks, and assessing ecosystem response to climate change.