Artificial Intelligence (AI) Use Case Inventory

By tracking current and planned Artificial Intelligence (AI) use cases – and sharing those with relevant internal and external audiences through Interior's Use Case Inventory – the Department will be able to identify and prioritize deployment of uses that are proven to enhance mission delivery, while also spurring the development of new use case ideas for development and testing. In addition to promoting the creation of new use cases, Interior will also work with partners within the Department, across the federal government, and in the academic, nonprofit, and private sectors to collaborate on adapting use cases to fit Interior’s mission. The Department does not currently have any AI use cases that are rights or safety impacting.   

Use Case Name Agency Bureau / Department Summary of Use Case Stage of System Development Life Cycle
Seasonal/Temporary Wetland/Floodplain Delineation using Remote Sensing and Deep Learning Department of the Interior BOR Reclamation was interested in determining if recent advancements in machine learning, specifically convolutional neural network architecture in deep learning, can provide improved seasonal/temporary wetland/floodplain delineation (mapping) when high temporal and spatial resolution remote sensing data is available? If so, then these new mappings could inform the management of protected species and provide critical information to decision-makers during scenario analysis for operations and planning. Completed
Data Driven Sub-Seasonal Forecasting of Temperature and Precipitation  Department of the Interior BOR Reclamation has run 2, year-long prize competitions where partisans developed and deployed data driven methods for sub-seasonal (2-6 weeks into future) prediction of temperature and precipitation across the western U.S. Participants outperformed benchmark forecasts from NOAA. Reclamation is currently working with Scripps Institute of Oceanography to further refine, evaluate, and pilot implement the most promising methods from these two competitions. Improving sub-seasonal forecasts has significant potential to enhance water management outcomes.  Development (not in production) 
Data Driven Streamflow Forecasting Department of the Interior BOR Reclamation, along with partners from the CEATI hydropower industry group (e.g. TVA, DOE-PNNL, and others) ran a year-long evaluation of existing 10-day streamflow forecasting technologies and a companion prize competition open to the public, also focused on 10-day streamflow forecasts. Forecasts were issued every day for a year and verified against observed flows. Across locations and metrics, the top performing forecast product was a private, AI/ML forecasting company - UpstreamTech. Several competitors from the prize competition also performed strongly; outperforming benchmark forecasts from NOAA. Reclamation is working to further evaluate the UpstreamTech forecast products and also the top performers from the prize competition.  Development (not in production) 
WOS.OS.NHM National Temperature Observations Department of the Interior USGS The objectives of this project are to reduce the burden on Science Centers for the collection, storage, analysis, and processing of quality assurance data with the expectation this will lead to an increase of deployed sensors in the water temperature network. More specifically the project will 
(1) modify software to allow for processing and storage of discrete water temperature data collected during streamflow measurements, 
(2) implement workflows and QA checks in data collection software that supports new temperature policies and procedures 
(3) create a pilot program to support Science Centers in accomplishing 5-pt temperature checks.		 Initiation
Snowcast Showdown  Department of the Interior BOR Reclamation partnered with Bonneville Power Administration, NASA - Goddard Space Flight Center, U.S. Army Corps of Engineers, USDA - Natural Resources Conservation Service, U.S. Geological Survey, National Center for Atmospheric Research, DrivenData, HeroX, Ensemble, and NASA Tournament Lab to run the Snowcast Showdown Prize Competition. In this competition, participants were asked to develop methods to estimate distributed snow information by blending observations from different sources   using machine learning methods that provide flexible and efficient algorithms for data-driven models and real-time prediction/estimation. Winning methods are now being evaluated and folded into a follow-on project with NOAA's River Forecast Centers.  Development and Acquisition
WRA.WPID.IWP.PUMP ExaSheds stream temperature projections with process-guided deep learning Department of the Interior USGS This 3-year project will improve PGDL stream temperature models by adding new forms of process guidance and merging techniques developed by USGS and DOE staff in past projects. Model assessments will emphasize robustness to projections in not-previously-seen conditions, such as those of future climates, paving the way for reliable projections into future decades in the Delaware River Basin. Implementation/Assessment
Vegetation and Water Dynamics  Department of the Interior USGS Major activities include tracking vegetation phenology as a basic input for drought monitoring and for capturing the unique phenological signatures associated with irrigated agriculture and invasive species. Drought mapping and monitoring focus on two conterminous US wide operational tools, VegDRI and QuickDRI, to inform drought severity in a timely fashion. A targeted livestock forage assessment tool is tailored to quantify drought effects in terms of livestock forage deficits in kg/ha for specific producer decision makers. High latitude systems have high carbon stocks, particularly the numerous wetlands. Understanding spatiotemporal surface water dynamics will inform of permafrost degradation and probable methane emission hot spots. Vegetation phenology signatures improve land cover class separations and capture unique phenological signatures associated with invasive species like cheatgrass. Understanding remote sensing sensitivity of phenology tracking at various spatial resolutions and varying degrees of noise associated with mixed pixel effects of other vegetation, soils, and water improves accuracy and consistency of estimations of phenology as well as derivative products tailored for specific land manager use. The determination of irrigated and non-irrigated system provides useful geospatial data for water management and can serve to isolate ecological comparisons or contrasts to either irrigated or non-irrigated land management. Operation and Maintenance
Data Mining, Machine Learning and the IHS Markit Databases Department of the Interior USGS To identify areas of potential conflict between energy development and alternative priorities, through the application of machine learning techniques to extract spatial patterns related to future development.

Lay the groundwork for the addition of new sets of skills, new types of analyses, and new products for the ERP and for the Mission Area; build internal knowledge about what machine learning can do for the ERP.		 Implementation
PyForecast  Department of the Interior BOR Pyforecast is a statistical/ML water supply forecasting software developed by Reclamation that uses a range of data-driven methods.  Implementation
DOMESTIC WELL VULNERABILITY SES INDICATORS NEW HAMPSHIRE Department of the Interior USGS The goals of this work are to: (1) investigate homeowner-level statistical associations between datasets on private wells (geology and land use, construction, hydraulics, and chemistry) and SES (and SES proxy) data; (2) investigate statewide census block-group level statistical associations between datasets on private wells (geology and land use, construction, hydraulics, and probabilities of arsenic and uranium contamination) and demographic and SES (and SES proxy) data; (3) identify indicators or triggers of vulnerability to private well water availability and quality in New Hampshire; and (4) broadly disseminate information from this study to scientific and general audiences, as well as to targeted community groups.  Implementation/Assessment
Aluminum Criteria Development in Massachusetts Department of the Interior USGS The USGS, in cooperation with MassDEP, will collect water-quality data at freshwater sites in Massachusetts, and use those data to demonstrate a process for calculating aluminum criteria based on a sites water chemistry (pH, DOC, and hardness) using a multiple linear regression model developed by the EPA (2017). Implementation/Assessment
TMDL and Data Mining Investigations Department of the Interior USGS Apply data-mining techniques, include artificial neural network models, the hydrologic investigations. Operation and Maintenance
Two-Dimensional Detailed Hydraulic Analysis Department of the Interior USGS The USGS proposes to conduct analysis of detailed hydrology and develop a two-dimensional hydraulic model to assist in decision-making for the protection of life and property and local floodplain management and regulation. The following objectives are identified in the scope of this effort:
Data objectives include:
1. Topographic surveys in the study reaches to verify or augment existing topography used in prior analyses.
a. Transportation routes
b. Critical infrastructure
c.  Various landforms
d. Non-structural flood mitigation recommendations at specific asset locations(USACE, 2019)
Interpretive objectives include:
1. Hydrologic analysis of the main stem of Joachim Creek (fig. 2) to produce discharge- frequency values for the 10%, 4%, 2%, 1%, and 0.2% regulatory flood flows.
2. Develop a calibrated two-dimensional hydraulic model inclusive of the following study reaches for the newly developed regulatory flood flows identified in interpretive objective (1) above:
a. Main stem 3.5-mile reach of Joachim Creek from a location above Highway Edownstream to cross-section AI (fig. 2). The study reach is aligned with the existing regulatory FIS effective model and FIRM bounded upstream at a mid- point location between cross-section BC and BB and downstream at cross- section AI (fig. 3a, 3b).
3. Two-dimensional model simulations of 10%, 4%, 2%, 1%, and 0.2% regulatory flood flows developed in interpretive objective (1) will produce flood profiles for the mainstem of Joachim Creek.
4. Development of two-dimensional model-derived flood maps for the main stem of Joachim Creek, will be disseminated for the newly defined 1% and 0.2% regulatory flood flows in interpretive objective (1). Model-derived maps will illustrate inundation extents, water-surface elevation, depth, and velocity, including a published table of comparisons with the summarized list of spatially relevant nonstructural flood mitigation assets defined in the preliminary FMP by USACE (USACE, 2019).		 Initiation
GEMSC Geospatial Modernization and Machine Learning Integration  Department of the Interior USGS The USGS Director's office laid out a vision for the USGS for the next decade in the blog post “21st Century Science—Preparing for the Future.” A key component of this vision was outlined by stating “Over the next decade, we will take advantage of advances in sensor technologies, integrated modeling, artificial intelligence (AI), machine learning (ML), and high-performance computing to observe, understand, and project change across spatial and temporal scales in real-time and over the long term.” For GEMSC to play a role in this initiative, a multi-year project is proposed to integrate these technologies in GEMSC project workflows and data services. The overarching objective for this project is development of a strategic framework for integrating ERP science with traditional information technology related platforms. Acquisition/Development
21st Century Prospecting:   AI-assisted Surveying of Critical Mineral Potential (Reimbursable) Department of the Interior USGS Based on the mandate to assess critical minerals distributions in the US, MRP has entered into a partnership between USGS and DARPA. The objective of this partnership is to accelerate advances in science for understanding critical minerals, assessing unknown resources, and increase mineral security for the Nation.  Acquisition/Development
WRA.HIHR.WAIEE Building capacity for assessment and prediction of post-wildfire water availability Department of the Interior USGS All listed objectives are focused on the western US:
· Collect multiple harmonized datasets from fire-affected basins in the western US that will advance development, calibration, and validation of water-quality models and assessment.
· Analyze harmonized datasets to assess regional differences in critical drivers of water quality impairment.
· Develop decision tree and standardized plan to determine locations to monitor after wildfire and ensure consistent post-fire water-quality data collection that accurately captures magnitude and duration of impairment.
· Develop rapid response plan to enable WSCs and WMA to be prepared for immediate responses for post-fire data collection and assessment.
· Establish the state of the science of critical drivers of post-fire water quality impairment in different ecoregions the western U.S.
· Characterize critical drivers, including in-stream and reservoir-sediment interface contributions, to post-fire water quality impairment.
· Build catalog of methods for measuring remotely sensed water quality after wildfire and apply multiple test cases of application.
· Develop catalog of critical data needs for geospatial prediction of wildfire impacts on water.
· Construct blueprint for incorporating missing critical water-quality impairment processes into modeling and prediction.
· Prepare plan with IWP for incorporating wildfire effects on water availability into rapid prediction.
· Participate in development and application of a framework for cross-Mission Area integration of predictive approaches spanning temporal and spatial scales for post-fire hazards.		 Initiation
Multi-scale modeling for ecosystem service economics Department of the Interior USGS Work continues to expand the existing ARIES modeling framework using artificial intelligence and a set of decision rules to build a system that can select models and data based on appropriate contextual factors (e.g., climate, vegetation, soils, socioeconomics). Using national and global datasets, this system will be capable of mapping ES at a much greater level of accuracy than before. The project will expand and implement this intelligent modeling system to the United States, yielding a consistent, nationwide, AI-supported intelligent ES modeling system to support ES assessment and valuation nationwide and beyond. This includes the integration of national economic accounts data with ecosystem services data to provide more timely, up to date, and integrated data at the national and subnational levels. Acquisition/Development, Implementation/Assessment
SWFL Habitat GIS Model Department of the Interior USGS Objective 1 – Update and maintain a seamless digital library of predicted flycatcher breeding habitat displayed (rendered) as binary or 5-class probability maps. This effort is ongoing. Landsat reimages the same location every 16 days. Currently, the digital library that is housed within ESRI’s AGOL library contains SWFL habitat maps from 2013 – 2022, spanning 57 Landsat scenes (see Hatten, 2016 for details) output by GEE.

Objective 2 – Update and maintain the SWFL Habitat Viewer so users can leverage and display the satellite model’s range-wide database and produce a habitat map for any stream reach in the flycatcher’s range. The web based (AGOL) application will allow one to query, 
display, and download flycatcher habitat maps from 2013 to present by leveraging a library of existing habitat maps generated with GEE, create a habitat time series for a given reach, produce a change detection map between two time periods, and produce metadata records based upon the scene’s date and digital footprint. The SWFL Habitat Viewer can also quantify or simulate beetle impacts to flycatcher habitat on a reach-by-reach basis, but simulations are dependent upon the availability of tamarisk maps. 

Objective 3 – Participate on regional workgroups, symposia, and conferences to inform potential and existing users about the SWFL Habitat Viewer. Currently, the RiversEdge West biannual conference and NAU’s biannual Colorado Plateau Research conference are the major outlets for presentations, but other regional conference candidates may be in Colorado, Nevada, New Mexico, or California.

Objective 4 - Collaborate in efforts to improve and extend the utility of the flycatcher satellite model by exploring cutting-edge modeling techniques (e.g., occupancy modeling, climate-wildlife modeling). For example, the flycatcher satellite model is being used to develop a regional database that contains patch attributes of SWFL habitat across the entire range of flycatchers. Such information is invaluable for exploring the relationships between patch occupancy and neighborhood characteristics (e.g., number of patches within a given radii, age of patches, distance between patches). The SWFL model is also being integrated into a regionwide project that focuses on linking interdisciplinary scientific data and models with artificial intelligence techniques, with a focus on hydrologic and ecological model integration in the Colorado River Basin, to better address drought and climate change.		 Initiation
Twitchell Rice AFRI Department of the Interior USGS A large, interdisciplinary study (led by UC Davis in collaboration with UC Berkeley, the USGS and several private consultants) will be investigating the effects (subsidence, gas flux and water quality) of converting acreage on Twitchell Island, a deeply subsided island in the Sacramento-San Joaquin Delta, from drained row crops to flooded rice production. The USGS research objective is to assess water quality effects with respect to MeHg production under different rice management practices including tillage, flooding and fertilization quantifying the relative methylation potential of each practice. Implementation
Kaguya TC DTM Generation Department of the Interior USGS The primary goals for FY21 are to develop a processing pipeline for generating Kaguya TC DTMs, generate a test suite of 100 Kaguya TC DTMs using Ames Stereo Pipeline (ASP), and evaluate the resulting products. Initiation
AI/ML for aquatic science Department of the Interior USGS This project aims to develop novel computational frameworks and AI algorithms for individual
fish recognition, by leveraging AI, computer vision and deep learning. The main objectives of
this project include:
(1) Develop baseline AI models by exploiting visual features and pre-trained deep
learning models.
(2) Improve individual fish recognition performance, as well as handling new individuals and
exploring dynamic environments.
(3) Evaluate melanistic markings associated with “blotchy bass syndrome” to assess the
capacity for AI detection of diseased fish.
(4) Evaluate deep learning models for individual recognition and respiration rate (ventilate
rate) using video data collected in laboratory settings and natural streams.		 Initiation
WRA.NWC.WU Gap analysis for water use Department of the Interior USGS

The USGS Water Use Program requires a formal and detailed gap analysis of water-use data for the nation in order to better understand uncertainty in water-use estimates and to help inform future data collection and modeling efforts. The primary objectives of this project are to: 
1) identify the dominant water-use categories in different areas of the U.S.; 
2) identify gaps in the available data for those categories, primarily gaps in data that if filled will improve model performance; and 
3) identify potential methods for data estimation that can be used to fill gaps and provide the most benefit to water-use modeling efforts. 

Other objectives include:  
1) increasing understanding of data quality to help inform uncertainty in model predictions; 2) collaboration with model developers to understand water-use model sensitivity to input data in order to focus and prioritize future data collection; and 
3) improved quality of data related to the extraction, delivery, and consumptive use of water for the important water use categories in different regions. Water-use categories include public supply, domestic, industrial, thermoelectric power, irrigation, livestock, and aquaculture. National models currently are under development for public supply, irrigation and thermoelectric. 

 Operations/Maintenance
WRA.NWC.IWAA National Extent Hydrogeologic Framework for NWC Department of the Interior USGS The primary objectives of this project are to 
(1) provide Nationally consistent predictions of groundwater quality (salinity and nutrients) relevant for human and ecological uses and its influence on surface-water, and 
(2) develop strategies for integrating these predictions into comprehensive water-availability assessments including the National Water Census and regional Integrated Water Availability Assessments. 
These primary objectives are organized by task as follows:
Task 1: Groundwater-Quality Prediction – salinity
· Provide accurate and reliable predictions of groundwater salinity at appropriate resolutions to document groundwater availability for human and ecological uses.
Task 2: Groundwater-Quality Prediction – nutrients
· Provide accurate and reliable predictions of nutrient concentrations in groundwater at appropriate resolutions to document groundwater availability for human and ecological uses.
Task 3: Incorporate Groundwater-Quality Predictions into Comprehensive Assessments of Water Availability
· Develop and refine strategies for coupling predictions of groundwater quality with groundwater flow and flux simulations from process-based models (e.g., GSFLOW, General Simulation Models) to quantify the amount of groundwater of a specified quality that is available and to better determine the affect of groundwater on surface-water quantity and quality.		 Initiation
WRA.NWC.IWAA National-Extent Groundwater Quality Prediction for the National Water Census and Regional Integrated Water Availability Assessments Department of the Interior USGS The primary objectives of this project are to (1) provide Nationally consistent predictions of groundwater quality (salinity and nutrients) relevant for human and ecological uses and its influence on surface-water, and (2) develop strategies for integrating these predictions into comprehensive water-availability assessments including the National Water Census and regional Integrated Water Availability Assessments.  Implementation/Assessment
WRA.HIHR.WQP Process-guided Deep Learning for Predicting Dissolved Oxygen on Stream Networks Department of the Interior USGS The objective of this project is to build a model that predicts daily minimum, mean, and maximum stream DO levels on stream segments in the Lower Delaware River Basin using nationally available datasets. Operations/Maintenance
WRA.NWC.EF Economic Valuation of Ecosystem Services in the Delaware River Basin Department of the Interior USGS The objectives of this project are to:   

Create a data and model inventory plan to evaluate existing data and models. 

Develop a database for the existing fish data. 

Develop Artificial Intelligence/Machine Learning (AI/ML) models to predict fish abundances and size under alternate future climates and reservoir operations. 

Develop models for economic valuation of the fishery resource. 

Evaluate the validity of estimated economic models against alternative approaches. 

Link models together to allow evaluation of tradeoffs between water use and the fisheries resource. 

Provide a prototype web application with re-usable components for internal USGS use that promotes understanding of the models and allows assessment of resource tradeoffs. 		 Initiation
WRA.NWC.IWAA Model Application for the National IWAAs and NWC Department of the Interior USGS In support of both the periodic National Water Availability Assessment reports and the routinely updated National Water Census, the Model Application for the National IWAAs and NWC (MAPPNAT) project will have four major objectives related to model application development: 
1) Provide initial applications of models for the National IWAAs reports and the National Water Census, 
2) Provide periodic long-term projections for the National IWAAs reports and the National Water Census, 
3) Provide routine model updates of current or near-current conditions for the National IWAAs reports and the National Water Census, and 
4) Provide operational short-term forecasts for the National Water Census. These four objectives will ultimately cover multiple hydrologic sub-disciplines—including water budgets, water use, water quality, aquatic ecosystems, and drought. Objective 1 will require a combination of on-project and off-project modeling activities to provide the needed model applications for National IWAAs and NWC version 1. Objectives 2, 3, and 4 will begin with strategic planning activities before implementation using the available model applications. As new models are developed, the staffing, organization, and approach for this project will be developed in an integrated manner that can accommodate multiple sub-disciplines and differing domain expertise. 		 Initiation
Improved Processing and Analysis of Test and Operating Data from Rotating Machines Department of the Interior BOR This project is exploring a better method to analyze DC ramp test data from rotating machines. Previous DC ramp test analysis requires engineering expertise to recognize characteristic curves from DC ramp test plots. DC ramp tests produce a plot of voltage vs current for a ramping voltage applied to a rotating machine. By using machine learning/AI tools, such as linear regression, the ramp test plots can be analyzed by computer software, rather than manual engineering analysis, to recognize characteristic curves. The anticipated result will be faster and more reliable analysis of field-performed DC ramp testing. Investigating/Proof of concept
WRA.WPID.IWP.PUMP Turbidity Forecasting Department of the Interior USGS This project aims to advance the use of national hydrological forecast models for delivering water quality forecasts relevant to water resource managers.  Implementation/Assessment
Improving UAS-derived photogrammetric data and analysis accuracy and confidence for high-resolution data sets using artificial intelligence and machine learning Department of the Interior BOR UAS derived photogrammetric products contain a large amount of potential information that can be less accurate than required for analysis and time consuming to analyze manually. By formulating a standard reference protocol and applying machine learning/artificial intelligence, this information will be unlocked to provide detailed analysis of Reclamation's assets for better informed decision making. Proof-of-concept completed
Photogrammetric Data Set Crack Mapping Technology Search  Department of the Interior BOR This project is exploring a specific application of photogrammetric products to process analysis of crack mapping on Reclamation facilities. This analysis is time consuming and has typically required rope access or other means to photograph and locate areas that can now be reached with drones or other devices. By formulating a standard reference protocol and applying machine learning/AI, this information will be used to provide detailed analysis of Reclamation assets for better decision making.  Proof-of-concept completed
Generative AI to Improve Visitor Experience on NPS.gov Department of the Interior NPS This is a proof of concept to us generative AI to extract data from NPS.gov and the NPS API to bring forth relevant content to visitors based on topics of interest allowing for improved trip planning. This proof of concept enriches structured data without requiring parks to create new content and reduced the time and labor cost of re-creating content.  Proof-of-concept completed

View the full 2023 AI Use Case Inventory.

Was this page helpful?

Please provide a comment