Skip to main content
  1. ScienceBase Catalog
  2. USGS Lower Mississippi-Gulf Water Science Center
  3. Modeled daily salinity derived from multiple machine learning methodologies and generalized additive models for three salinity monitoring sites in Mobile Bay, northern Gulf of Mexico, 1980–2021

Modeled daily salinity derived from multiple machine learning methodologies and generalized additive models for three salinity monitoring sites in Mobile Bay, northern Gulf of Mexico, 1980–2021

Dates

Publication Date
Start Date
1980-01-08
End Date
2021-12-31

Citation

Banks, S.M., DeAbreu, C.J.P., Roland, V.L., Asquith, W.H., and Rodgers, K.D., 2024, Modeled daily salinity derived from multiple machine learning methodologies and generalized additive models for three salinity monitoring sites in Mobile Bay, northern Gulf of Mexico, 1980–2021: U.S. Geological Survey data release, https://doi.org/10.5066/P9NIIEJJ.

Summary

Results from generalized additive models (GAM), random forest models (RFM), and cubist models (CUB) for three Dauphin Island Sealab (DIS) operated salinity sites in Mobile Bay are reported in this data release. These sites included Meaher Park (DIS:MHPA1), Middle Bay Lighthouse (DIS:MBLA1), and Dauphin Island (DIS:DPIA1). The constructed models predicted a 42-year daily salinity record from 1980 to 2021 at each site based on incomplete imputed salinity records and several explanatory variables. Explanatory variables included: daily streamflow from 8 United States Geological Survey (USGS) streamgages, daily minimum and maximum temperature, precipitation, vapor pressure, wind speed, wind direction, horizontal and vertical wind speed [...]

Contacts

Point of Contact :
William H Asquith
Process Contact :
Kirk D Rodgers
Originator :
Sarah M Banks, Caleb JP DeAbreu, Victor L Roland, William H Asquith, Kirk D Rodgers
Metadata Contact :
Sarah M Banks
Publisher :
U.S. Geological Survey
USGS Mission Area :
Water Resources
SDC Data Owner :
Lower Mississippi-Gulf Water Science Center
Distributor :
U.S. Geological Survey - ScienceBase

Attached Files

Click on title to download individual files attached to this item.

Modeled daily salinity for three salinity monitoring sites in the Mobile Bay northern Gulf of Mexico.xml
Original FGDC Metadata
View 		411.15 KB application/fgdc+xml
CUB_DIS_DPIA1_predictions.txt 1.65 MB text/plain
CUB_DIS_MBLA1_predictions.txt 1.62 MB text/plain
CUB_DIS_MHPA1_predictions.txt 1.55 MB text/plain
CUB_MOBAY_var_imp_summary.txt 6.89 KB text/plain
diagnostics_all_models.txt 3.01 KB text/plain
ENSEMBLE_DIS_DPIA1_predictions.txt 2.26 MB text/plain
ENSEMBLE_DIS_MBLA1_predictions.txt 2.19 MB text/plain
ENSEMBLE_DIS_MHPA1_predictions.txt 2.05 MB text/plain
GAM_DIS_DPIA1_predictions.txt 1.65 MB text/plain
GAM_DIS_MBLA1_predictions.txt 1.61 MB text/plain
GAM_DIS_MHPA1_predictions.txt 1.56 MB text/plain
Mississippi Sound Salinity sites.txt 2.77 KB text/plain
Mississippi Sound Streamflow sites.txt 929 Bytes text/plain
MOBAY_model_input.txt 26.95 MB text/plain
RFM_DIS_DPIA1_predictions.txt 1.66 MB text/plain
RFM_DIS_MBLA1_predictions.txt 1.62 MB text/plain
RFM_DIS_MHPA1_predictions.txt 1.55 MB text/plain
RFM_MOBAY__var_imp_summary.txt 6.97 KB text/plain

Purpose

The purpose of this data release is to document the input and output of Cubist and Random forest machine learning models, and generalized additive models developed to simulate day-to-day variability in salinity predictions.

Map

Map Preview

Spatial Services

ScienceBase WMS

Communities

  • USGS Data Release Products
  • USGS Lower Mississippi-Gulf Water Science Center

Tags

Provenance

Data source
Input directly
Landing page and metadata updated by Daniel Kroes on June 4, 2024. The updates include switching "Generalized Additive" to "generalized additive", "Random Forest" to "random forest", "Cubist Models" to "cubist models", and "Model" to "model" within the title, citation, and abstract. The third sentence in the abstract was also updated from "The constructed models predicted a 40-year daily salinity record..." to "The constructed models predicted a 42-year daily salinity record...".

Additional Information

Identifiers

Type Scheme Key
DOI https://doi.org/10.5066/P967LATZ doi:10.5066/P9NIIEJJ

Item Actions

View Item as ...

JSON ATOM ISO XML

Save Item as ...

JSON FGDC MODS XML ISO XML

View Item...

Metrics