Checking for non-preferred file/folder path names (may take a long time depending on the number of files/folders) ...

Estimation of CONUS stream water quality using LSTM and WRTDS


Authors:
Owners: This resource does not have an owner who is an active HydroShare user. Contact CUAHSI (help@cuahsi.org) for information on this resource.
Type: Resource
Storage: The size of this resource is 5.0 GB
Created: Aug 15, 2024 at 3:55 p.m.
Last updated: Sep 03, 2024 at 5:21 p.m.
DOI: 10.4211/hs.8da6ebf2ee9a491490bb09a6349e70fe
Citation: See how to cite this resource
Sharing Status: Published
Views: 158
Downloads: 10
+1 Votes: Be the first one to 
 this.
Comments: No comments (yet)

Abstract

This dataset offers a comprehensive collection of water quality data for approximately 500 stations across the Continental United States (CONUS). It includes 20 common water quality parameters, along with meteorological, hydrological, and land use variables such as streamflow, precipitation, temperature, evapotranspiration, and vegetation indices. To support water quality modeling research, we provide model outputs from both conventional statistical (WRTDS) and advanced deep learning (LSTM) approaches. This dataset is designed to facilitate model development, validation, and application, and to promote reproducible research.

Subject Keywords

Coverage

Spatial

Coordinate System/Geographic Projection:
WGS 84 EPSG:4326
Coordinate Units:
Decimal degrees
Place/Area Name:
continental US
North Latitude
50.0000°
East Longitude
-65.0000°
South Latitude
24.0000°
West Longitude
-125.0000°

Temporal

Start Date:
End Date:

Content

readme.txt

# Introduction
This dataset contains water quality data collected from approximately 500 stations across the Continental United States (CONUS). The data includes measurements of 20 water quality parameters and corresponding meteorological, hydrological, and land use data. The dataset aims to support research and modeling efforts related to water quality.

# Data Sources
Streamflow and water quality data: USGS
Daily climate forcing: gridMET
Chemical composition of precipitation: National Trends Network (NTN)
Daily remote sensing vegetation indexes: GLASS

# Data Structure
The data is organized into two primary folders:

## stations
This folder contains individual CSV files for each station of size [#time, #variables], with each file covering the period from 1982-01-01 to 2018-12-31. Each row represents a day, and each column contains a specific variable.

- streamflow from USGS: 
  - streamflow: Streamflow in cubic meters per second (m3/s)
  - runoff: Runoff in millimeters (mm)

- Daily climate forcing from gridMET
  - pr: Precipitation in millimeters (mm)
  - sph: Specific humidity at 2 meters (kg/kg)
  - srad: Surface downward shortwave radiation (W/m²)
  - tmmn: Minimum temperature at 2 meters (degrees Celsius)
  - tmmx: Maximum temperature at 2 meters (degrees Celsius)
  - pet: Potential evapotranspiration (mm)
  - etr: Actual evapotranspiration (mm)

- Chemical composition of precipitation from the National Trends Network (NTN)
  - ph: pH
  - Conduc: Conductivity (µS/cm)
  - Ca: Calcium concentration (mg/L)
  - Mg: Magnesium concentration (mg/L)
  - K: Potassium concentration (mg/L)
  - Na: Sodium concentration (mg/L)
  - NH4: Ammonium concentration (mg/L)
  - NO3: Nitrate concentration (mg/L)
  - Cl: Chloride concentration (mg/L)
  - SO4: Sulfate concentration (mg/L)
  - distNTN: Distance to the nearest nutrient sampling site (km)

- Daily remote sensing vegetation indexes from GLASS
  - LAI: Leaf Area Index
  - FAPAR: Fraction of Absorbed Photosynthetically Active Radiation
  - NPP: Net Primary Productivity

- Others
  - datenum: Date in datenum format
  - sinT: Sine of time
  - cosT: Cosine of time
    

## results
This folder contains the model outputs in CSV format. Each file  are of size [#time, #site], has the following structure:

- LSTM_{code}.csv: LSTM predictions for water quality parameter with code {code}.

- WRTDS_{code}.csv: WRTDS predictions for water quality parameter with code {code}.

- Obs_{code}.csv: Observed values for water quality parameter with code {code}.

- streamflow.csv: Streamflow data for all stations.

- maskTrain.csv: Training mask indicating training periods.

- maskTest.csv: Testing mask indicating testing periods.
#Contact
  For questions or issues, please contact Kuai Fang kuaifang@stanford.edu.

Credits

Funding Agencies

This resource was created using funding from the following sources:
Agency Name Award Title Award Number
U.S. Department of Energy DE-SC0018155
Stanford University Human-Centered AI (HAI) program and Data Science fellowship

Contributors

People or Organizations that contributed technically, materially, financially, or provided general support for the creation of the resource's content but are not considered authors.

Name Organization Address Phone Author Identifiers
Kate Maher

How to Cite

Fang, K. (2024). Estimation of CONUS stream water quality using LSTM and WRTDS, HydroShare, https://doi.org/10.4211/hs.8da6ebf2ee9a491490bb09a6349e70fe

This resource is shared under the Creative Commons Attribution CC BY.

http://creativecommons.org/licenses/by/4.0/
CC-BY

Comments

There are currently no comments

New Comment

required