Cooperative Ecosystem Studies Unit, Rocky Mountain CESU | G17AS00049

Frequently Asked Questions (FAQs)

What is the Rocky Mountain CESU funding opportunity (G17AS00049) about?

This Cooperative Ecosystem Studies Unit (CESU) funding opportunity supports a US Geological Survey (USGS) Northern Rocky Mountain Science Center (NOROCK) cooperative agreement focused on improving how ecologists combine and analyze complex, mixed ecological data. The central problem it addresses is that modern ecological research and management often require integrating heterogeneous information collected using different methods, across different places, time periods, and species. The opportunity emphasizes that incorrect or overly simplistic data integration can lead to biased results and uncertainty that is underestimated or not recognized.

Which agency and office are sponsoring the work?

The opportunity is sponsored by the US Geological Survey (USGS), specifically the Northern Rocky Mountain Science Center (NOROCK), and it is run under the Cooperative Ecosystem Studies Unit (Rocky Mountain CESU) framework.

What type of funding instrument is being used?

The funding instrument is a cooperative agreement under the Department of the Interior (DOI), USGS. The opportunity is described as discretionary funding.

What is the CFDA number for this opportunity?

The CFDA number listed for this opportunity is 15.808.

How much funding is available, and is there an award ceiling?

The opportunity anticipates a single award with an award ceiling of $40,118.

How many awards are anticipated?

The announcement anticipates a single award.

Who is eligible to apply?

Eligibility is limited to CESU partners, consistent with typical CESU announcements that leverage a network of partner institutions. The specific eligibility details are referenced in the opportunity's eligibility clarification text.

What is the main goal of the funded project?

The funded work is aimed at producing a rigorous review of Bayesian approaches for integrating multi-type and multi-species ecological data, with a strong emphasis on spatial and spatiotemporal settings. The project is intended to clarify what methods work, what assumptions are commonly made, and where current approaches fall short when data have spatial and temporal dependence.

Is the project primarily about collecting new field data?

No. The described focus is on synthesizing and evaluating existing statistical methods and the research literature. The emphasis is on a review and technical synthesis rather than new field data collection.

What kinds of data integration challenges is the opportunity trying to address?

The opportunity targets the challenge of integrating heterogeneous ecological datasets that differ by collection method, spatial extent, time period, and species coverage. It highlights that many real-world ecological questions cannot be answered well using a single "tidy" dataset, and that poor integration can create bias and lead to uncertainty being understated or overlooked.

Why is there a strong emphasis on spatial and spatiotemporal settings?

NOROCK emphasizes spatial and spatiotemporal settings because ecological processes are rarely independent across space and time. Nearby locations tend to be related, and repeated observations through time are also related. The review is expected to treat these dependencies as central to the modeling problem rather than as secondary considerations.

What statistical framework is the review expected to focus on?

The review is expected to focus on Bayesian approaches, particularly Bayesian hierarchical models, for integrating complex ecological datasets in spatial and spatiotemporal contexts.

What are Integrated Population Models (IPMs), and why are they highlighted?

Integrated Population Models (IPMs) are highlighted as a structured way to combine multiple datasets (for example, counts, capture-recapture, productivity, telemetry, or occupancy-type observations) to estimate demographic or population parameters more effectively than any single dataset could. The opportunity emphasizes their practical value because ecological data are often expensive and difficult to collect and may be incomplete, making it important to blend information sources that vary in quality, scale, and sampling design.

What limitations of conventional IPMs does the opportunity want the review to address?

The opportunity notes that conventional IPMs often assume different datasets are independent. That assumption may be unrealistic when datasets share sampling sites, observers, detection processes, or underlying environmental drivers. The review is expected to address these issues, including how Bayesian methods can relax unrealistic independence assumptions and better represent shared processes and observation error.

What does the opportunity mean by "multi-type" ecological data?

Based on the description, "multi-type" refers to heterogeneous data sources collected with different methods and designs (such as counts, capture-recapture, productivity data, telemetry data, and occupancy-type observations), potentially spanning different spatial extents and time periods.

What does the opportunity mean by "multi-species" data, and why is it challenging?

Multi-species data involve observations or datasets that include more than one species. The opportunity emphasizes that analyses can be overly simplistic if they fail to account for species-to-species differences in detection, habitat relationships, or demographic rates. These shortcomings can lead to inaccurate inferences and uncertainties that are not properly identified.

What modeling expectations are described for multi-species analyses?

The review is expected to clarify how multi-species Bayesian hierarchical models can represent species heterogeneity while still borrowing strength across species when appropriate. It also underscores that spatial and spatiotemporal contexts can increase computational and modeling complexity, which the review should take into account.

What kinds of datasets are mentioned as examples that might be integrated?

The opportunity mentions examples including counts, capture-recapture, productivity, telemetry, and occupancy-type observations.

What is the expected deliverable?

The expected deliverable is a rigorous review that goes beyond summarizing papers. It should evaluate existing Bayesian methods for integrating multi-source and multi-species ecological data in spatial and spatiotemporal settings, identify common assumptions and their risks, describe gaps and limitations in current methods, and outline promising directions.

How is the review intended to be used after completion?

The review is intended to serve as a bridge to future work and proposals with other partners, particularly follow-on proposals that advance integrated population models (IPMs) and Bayesian hierarchical modeling approaches for multi-species and multi-source datasets.

What is the Funding Opportunity Number for this announcement?

The Funding Opportunity Number is G17AS00049.

When was the announcement created, and what was the closing date?

The announcement was created on March 29, 2017, and the original closing date was April 14, 2017.

What key risks or pitfalls in data integration does the opportunity highlight?

The opportunity highlights that missteps in combining datasets can produce biased conclusions and uncertainty that is underestimated or not recognized. It also emphasizes that assuming independence among datasets (when they share sites, observers, detection processes, or drivers) can be unrealistic and lead to misleading inference.

What is the role of Bayesian methods according to the opportunity description?

Bayesian methods are positioned as a way to integrate heterogeneous ecological datasets while better representing shared processes, observation error, and spatial/spatiotemporal dependencies, and by allowing models to relax unrealistic independence assumptions that appear in more conventional approaches.