Data Anomaly Detection and Sediment Yield Estimation in the US Army Corps of Engineers’ Reservoir Sedimentation Information (RSI) Database | W81EWF 20 SOI 0025

Opportunity Information: Apply for W81EWF 20 SOI 0025

This grant opportunity, titled "Data Anomaly Detection and Sediment Yield Estimation in the US Army Corps of Engineers' Reservoir Sedimentation Information (RSI) Database" (Funding Opportunity Number W81EWF 20 SOI 0025), is a Department of Defense research effort sponsored by the US Army Corps of Engineers. It is offered as a discretionary science and technology research award through a cooperative agreement, with an anticipated single award up to $40,000. The goal is practical and applied: improve the reliability and usefulness of the Corps' RSI database by finding bad or inconsistent records and, beyond that, using the cleaned and enriched information to estimate reservoir sedimentation rates more effectively.

The primary objective is focused on data quality assurance for the RSI system. The Corps is looking for a method that can identify erroneous or suspicious entries in the RSI dataset, ideally using machine learning-based anomaly detection rather than only manual review or simple rule checks. In other words, the work should help distinguish normal, plausible reservoir sedimentation records from records that look incorrect due to data entry mistakes, unit mismatches, missing or mis-typed values, inconsistent timestamps, or values that do not align with the physical behavior expected of sedimentation processes. The emphasis on anomaly detection implies the solution should be able to flag outliers and inconsistencies in a repeatable, scalable way, and not just fix a few known problems one time.

A secondary objective expands the project from data validation into predictive modeling. Using RSI data together with supplemental datasets from other sources, the investigator(s) are expected to develop a machine learning approach for estimating sedimentation rates (or sedimentation-related measures) at reservoirs. This part of the effort is about extracting more value from the RSI system by linking it with environmental, hydrologic, geographic, watershed, or operational variables that help explain sediment deposition over time. The intended outcome is a method that can make reasonable sedimentation-rate estimates where observations are sparse, inconsistent, or unavailable, while still being grounded in the best available composite dataset.

The opportunity lays out a set of research tasks that define the expected workflow. First, the project should identify appropriate supplemental data sources that can be integrated with RSI records. While the notice does not prescribe which sources must be used, the task implies finding datasets that add explanatory power and context, such as watershed characteristics, land cover and land use, soil and geology, precipitation and runoff indicators, streamflow records, reservoir geometry and capacity, survey histories, or other relevant federal or public datasets. Second, the investigator(s) should examine the RSI data to identify patterns and trends, which could include temporal trends in sedimentation rates, geographic clustering, differences across reservoir types or regions, relationships between survey intervals and reported sedimentation, or systemic issues that create recurring data problems (for example, recurring unit conversion errors or survey method artifacts). Third, using the combined dataset, the team should develop a machine learning method to detect anomalies in RSI, meaning a model-driven way to flag records that are statistically unusual or physically implausible when viewed in context with related variables. Fourth, the team should develop a machine learning method to estimate reservoir sedimentation rates, effectively building a predictive model that can generalize from existing RSI observations and the supplemental predictors.

Administratively, this is a cooperative agreement rather than a standard contract, which typically signals that the agency expects collaboration, interaction, or iterative feedback during the project period, even if the technical work is largely performed by the recipient. The funding is relatively modest (award ceiling $40,000) with one expected award, indicating a focused, proof-of-concept or pilot-scale effort rather than a multi-year, large-team program. The CFDA listing provided is 12.630, and eligibility is described broadly as "Others" with further clarification referenced in the full posting. The opportunity was created June 22, 2020, with an original closing date of August 14, 2020, placing it in a timeframe where the Corps was likely seeking targeted research support to modernize the RSI database workflow and analytics.

Taken together, the grant is essentially asking for two tightly related deliverables: a defensible, machine-learning-driven process to improve RSI data integrity through anomaly detection, and a companion predictive modeling approach that uses RSI plus external data to estimate sedimentation rates. The underlying theme is making reservoir sedimentation information more dependable for decision-making by both cleaning the data pipeline and building better estimation tools that can operate even when direct measurements are imperfect or incomplete.

• The Department of Defense, Dept. of the Army -- Corps of Engineers in the science and technology and other research and development sector is offering a public funding opportunity titled "Data Anomaly Detection and Sediment Yield Estimation in the US Army Corps of Engineers’ Reservoir Sedimentation Information (RSI) Database" and is now available to receive applicants.
• Interested and eligible applicants and submit their applications by referencing the CFDA number(s): 12.630.
• This funding opportunity was created on Jun 22, 2020.
• Applicants must submit their applications by Aug 14, 2020. (Agency may still review applications by suitable applicants for the remaining/unused allocated funding in 2026.)
• Each selected applicant is eligible to receive up to $40,000.00 in funding.
• The number of recipients for this funding is limited to 1 candidate(s).
• Eligible applicants include: Others (see text field entitled Additional Information on Eligibility for clarification).

Apply for W81EWF 20 SOI 0025

[Watch] Creating a grant proposal using the step-by-step wizard inside the applicant portal: