Relevant datasets include:
...
There are multiple types of water quality data, with varying nuances about why this data is collected (i.e., purpose) and who does the data collection. A brief description of different water quality data and notes re: these nuances are captured below.
Drinking water quality
NMED (SDWIS - captures 1000+ PWS)
Theoretically have the machine readable data available
SDWIS accessible online but not query-able per criteria (only viewable by well)
Connection between SDWIS and Drinking Water Watch (both EPA-driven) is somewhat locked under EPA set up
Surface water quality (SWQ)
NMED regulates surface water quality under the Clean Water Act (
CWA).
Approximately 700
water bodies assessed by NMED (for various
considers agricultural water quality issues
groundwater quality
petroleum storage tank monitoring wells
produced water quality
criteria)
Potentially have machine-readable data available
Most data are accessible through the USGS and EPA Water Quality Portal
Considers many categories and criteria, such as cold water aquatic life, recreational contact, and irrigation uses.
USGS collects intermittent and continuous surface water quality for research (on NWIS)
NMBGMR collects intermittent surface water quality for research
Universities collect intermittent SWQ for research
NGOs also collect SWQ
Groundwater quality
NMED regulates protection of groundwater from discharges
Mostly spreadsheets and some paper files
Monitoring well data - as required for discharge permits (such as petroleum storage tanks, dairy farms, or landfills) * these data include measurements of water levels
NMBGMR collects groundwater quality for hydrogeologic research
SQL database, can be made machine readable
USGS collects groundwater quality for hydrogeologic research
Produced water quality
Not regulated right now by NMED, no state agency database
Eventually PRRC/NMPWRC hope to be intaking some quality data for treated produced water (that’s used outside of O&G)
Precipitation water quality
Occasionally collected for research purposes (NMBGMR)
The table below summarizes how some of the existing water quality datasets could be mapped onto the SensorThings framework.
Table 1: The STA data model definitions and example mappings to NM agency datasets regarding water quality. Note: text in red indicates content actively being discussed.
SensorThings Entity | Description | Chemistry Surface Water Data (e.g.,
) | Groundwater Data (e.g., report generated from dtb:
) | Biological Data (e.g., CKAN dataset) | ||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Metadata | Location | A unique coordinate or area on the surface of the earth | Street Address (possibly with associated latitude and longitude). (e.g. 3960 PRINCE ST) - geoCoding required | Coordinates - SiteID | sys_loc_code - also has x,y,z coordinate (have it in both state plane (US survey foot) and wgs84) | Coordinates - everything’s recorded in UTM; also a lat-long uploaded with it | ||||||||||||||||
Thing | Some real-world entity with which one or more Sensors are associated | Sample Pt RT236I | SiteID WaterBody ID | sys_loc_code (e.g., LAF-24) | usually has a FieldID tied to a particular site, but does vary with sampling event | |||||||||||||||||
Datastream | A collection of Observations about an ObservedProperty produced by a Sensor associated with a Thing | Sample Results | Sample Results | report_result_value | Sample Results | |||||||||||||||||
Datastream/observationType | The type of observation, codified in the Observations and Measurements data standard. Types include Categorical (defined text), Count (integer), Measurement (continuous number), Observation (free text), and TruthObservation (True/False) | Categorical or TruthObservation multiDataStream if there are multiple analytes | Measurement (discrete values) | Measurement | Measurement | |||||||||||||||||
Datastream/unitOfMeasurement | A three-item definition of the unit of measurement, including its name, symbol, and link to the definition (preferably to one provided in an established ontology such as http://unitsofmeasure.org/ucum.html or http://qudt.org/) | TCR Result (binary indicator: absence/presence) ug/l (micrograms/liter; also = ppb), mg/l (also = ppm), nTu, deg C, %, us/cm (microSiemens/centimeter), “pH” (techically dimless), rads, | ug/l (micrograms/liter; also = ppb), mg/l (also = ppm), nTu, deg C, %, us/cm (microSiemens/centimeter), “pH” (techically dimless), rads, mpn (most probably number) | report_result_units ug/l, mg/l, % recovered (for surrogates), deg C, mg CO2/liter, mg CaCO3/liter, mV, nTu, pH, us/cm Sometimes rads | number of fish by species species identified by taxa code (6-letter code) sometimes temp (deg C) | |||||||||||||||||
Sensor | The procedure used to provide a Datastream. Can be a particular data recording device model, or a defined procedure followed by a human observer. If applicable, a specific instance (e.g. a sensor model and serial number) | 9223B-PA (https://www.standardmethods.org/doi/10.2105/SMWW.2882.194) | Lab results - 100+ methods Field results - sondes, data loggers | 2 methods fields: standard analytical method (~15 that are run) and a prep method | Highly variable - could be traps, net, electrofishing, … | |||||||||||||||||
ObservedProperty | The raw or processed phenomenon (quantitative or qualitative) being measured for the Datastream. Preferably including a link to a definition provided by an established ontology or controlled vocabulary such as the ODM2 Controlled Vocabularies or http://qudt.org/) | Analyte (e.g. Coliform (TCR) (3100)) Turbidity | Characteristic | cas_rn (could also use: chemical_name; there may be variants here) | Number | |||||||||||||||||
OPTIONAL: FeatureOfInterest | The real-world feature that the Observations are about. This may or many not be different from the Location where the Thing on which the Sensor is mounted. Can include a JSON-formatted point location or a polygon or collections thereof. | Public Water System (head office location or service area boundary) (e.g. Albuquerque Water System PWSID NM3510701) | Polygonal or Linear Feature - WaterBody ID Matrix code: W (for water) | screen interval/depth sample was collected at (z_from, z_to) Matrix code: W (for water) | HUC2, HUC8, Waterbody name | |||||||||||||||||
OPTIONAL: FeatureOfInterest | The real-world feature that the Observations are about. This may or many not be different from the Location where the Thing on which the Sensor is mounted. Can include a JSON-formatted point location or a polygon or collections thereof. | Qualifiers on observedproperty: 1) dilution value? 2) was water sample filtered (happens for select analytes, e.g., metals)? if so, at what microns? 3) detection limits 4) only ingesting normal samples (not dups, trip blanks, etc.) | 1) detection limits; quantitation (QDL), method detection (MDL), and reporting detection (RDL). Usually QDL and RDL are the same. 2) dilution factor 3) test type (some are “retest” if there’s an error in the lab) 4) sample filtering - D (xx), N (xx), and T (total) 5) fields for purge volume/time 6) lab id code 7) sample type code - N (normal), D (duplicate), … → only pull in Normal data? | |||||||||||||||||||
Data | Observation | A single measurement value including the result, time values, and other metadata. Information on the ObservedProperty that was measured by what Sensor is provided by the Datastream these observations are in. Features of Interest are linked for each observation as well. Observations are linked to (collected in) Datastreams | Sample (e.g. 763391) | Sample (~7 digits, e.g., 2555000) | 2 sample Ids: 1) CABQ (e.g., sys_sample_code:LAF24-20201110W-N) - trip blanks are also captured this way 2) lab sample id 3) test id | FieldID | ||||||||||||||||
Observation/result | The actual measured value, with valid values defined in observationType and units defined in unitsOfMeasurement, both provided by Datastream | Sample Result (P (Positive/ Coliform found) A (Negative/ Coliform not found)) | Sample Result | report_result_value | number | |||||||||||||||||
Observation/phenomenonTime | The date+time (or interval) in ISO 8601 format (YYYY-MM-DDT:HH:MM:SS-Z) when the observation occured | MP (Monitoring Period) (e.g. 01-01-2020 to 01-31-2020) (might need to discuss this further to figure out value for entry; does coincide with entry within Drinking Water Watch) | y-m-d h-m-s | sample_date (y-m-d h-m-s) | y-m-d | |||||||||||||||||
OPTIONAL: Observation/resultTime | The date+time that the result was generated. May be the same as phenomenonTime | Date (e.g., 01-06-2020) (refers to sampling time, not time of lab analysis) | also an analysis date + time that’s available: y-m-d h-m-s | also an analysis date + time that’s available: y-m-d h-m-s | ||||||||||||||||||
OPTIONAL: Observation/validTime | The date+time interval during which the Observation can be used (often used for provisional values that are replaced by QA/QC’d observations) | |||||||||||||||||||||
OPTIONAL: Observation/resultQuality | A description of the result Quality. Will vary according to agency practice. Can use ODM2 controlled vocabulary for data quality types as a guide. | 3 different fields: lab qualifier (describes matrix spikes, replicates, etc.); internal qualifier (result of validation process); STORET qualifier (translation of internal qualifier to federal code) | result_type_code: most are TRG, some are SUR (surrogates) 2 qualifiers: 1) detection flag - yes or no; 2) organic - yes or no also, have lab qualifiers (J, …) to show which detection bands are within, etc. | quality control checks involve data entry verifications (from std. data entry sheets into dtb). Post dtb verification, entry sheets are either archived in the office or at UNM. |