.. _gen_usgs_nwis: National Water Information Service ^^^^^^^^^^^^^^^^^^ Streamflow and groundwater data from the USGS National Water Information System (NWIS) database. - Daily streamflow and water table depth data are in the site's local time zone and obtained from the `"/daily" endpoint `_ of the updated USGS OGC API. - Hourly streamflow and water table depth data are in UTC and aggregated to the hourly level from the `"/continuous" endpoint `_ of the updated USGS OGC API. The raw observations are typically collected at 15-minute increments. - The water table depth data accessed with `temporal_resolution='instantaneous'` comes from the USGS `Groundwater Levels Service `_. Our team is currently working on updating this dataset to query from the updated USGS OGC endpoint. Note that these data usually do not have regular temporal coverage and many of the sites with data available through this method only have a single point-in-time observation available. **Dataset Name**: usgs_nwis **Data Source**: usgs **Data Collection or Processing Notes:** We query data from the USGS weekly, early on Sunday mornings. Each weekly job collects all observations that have been modified since the date of our previous data pull. For sites that are currently in operation, this translates to collecting data that was modified during only the previous week. Because of the sparsity of the `temporal_resolution='instantaneous'` groundwater measurements, those are not included in this weekly schedule. We plan to query that source for new observations roughly every few months. Note that raw hourly data is saved in UTC while raw daily data is saved with respect to the local site time zone. To maintain the integrety and traceability back to the original sources, our team conducts very limited data manipulation on the queried data. This includes the following: - Unit translation into SI units - Standardization of NaN/missing values - For example, USGS will sometimes provide strings such as "Ice" or "Dry" to indicate reasons for why certain observations are missing. A full list of such fields is available `here `_. We standardize these values into the numeric numpy.NaN to allow the entireity of the series to be interpreted as numeric. - Consolidating multiple concurrent data series - The USGS data sometimes provides multiple concurrent observation series for the same variable for the same site. In these cases, we consolidate the multiple series into a single series following these prioritizations: - If one of the series has been verified, we prioritize that over provisional data - If both series are identical values, we simply reduce down to a single set of observations - If one of the series has non-missing data and the other series has missing data, we prioritize the non-missing data - If multiple series remain with conflicting values, we take the average of the resulting non-missing values Variables ^^^^^^^^^^^^^^^^^^ This describes the available variables of the dataset. Use the dataset, variables and temporal_resolution in python access functions as described in the Working with Gridded Data, and Working with Point Observations. .. list-table:: Land Surface Variables :widths: 25 60 30 20 20 20 :header-rows: 1 * - variable - description - temporal_resolution - units - aggregation - 3D * - streamflow_anomaly - Streamflow percentile rank - weekly, monthly, daily - unitless - mean - no .. list-table:: Subsurface Variables :widths: 25 60 30 20 20 20 :header-rows: 1 * - variable - description - temporal_resolution - units - aggregation - 3D * - wtd_anomaly - Water table depth percentile rank - weekly, monthly, daily - unitless - mean - no * - water_table_depth - Water table depth - instantaneous, monthly, daily, weekly, hourly - m - mean - no .. list-table:: Surface Water Variables :widths: 25 60 30 20 20 20 :header-rows: 1 * - variable - description - temporal_resolution - units - aggregation - 3D * - streamflow - Streamflow - weekly, monthly, daily, hourly - m3/s - mean - no