Bat Acoustic Data Vetting: What It Is and Why ESA Compliance Depends on It

What acoustic bat surveys actually produce

When a bat biologist deploys ultrasonic detectors at a survey site, those detectors record echolocation calls throughout the night as bats fly through the detection zone. At the end of the survey period, what the biologist retrieves is a folder of audio files — typically full-spectrum WAV recordings — that may number in the hundreds or thousands depending on bat activity levels, survey duration, and site conditions.

Those files cannot simply be counted. Each one must be analyzed to determine whether it contains a bat call, whether it is of sufficient quality to be identified to species, and whether any of the recorded calls belong to a target species — Indiana bat (IBAT), northern long-eared bat (NLEB), or tricolored bat (TCB) — for ESA compliance purposes.



This analysis process involves two distinct stages under the 2026 guidelines: automated analysis using USFWS-approved software programs, followed — when warranted — by qualitative analysis, also referred to as visual vetting, in which a qualified biologist reviews the call files directly.

Stage one: automated analysis

The 2026 guidelines require surveyors to conduct automated acoustic analyses for all sites and all nights that were approved in the study plan. This is not optional. A survey that runs automated analysis only on nights when bats were active, or only on sites where the biologist expected detections, does not meet the standard.

Automated analysis runs each recorded file through a USFWS-approved software program that compares the statistical properties of the recorded call to a library of known calls. The program classifies each file to species and, when run at the site-night level (all files from a single detector location on a single night), produces a Maximum Likelihood Estimate — the MLE — which is a statistical measure of the probability that the target species is present at that site on that night.

The approved software programs — and their specific approved versions, settings, and recording formats — are listed on the USFWS Automated Acoustic Bat ID Software Programs webpage. Surveyors must use the most current approved software versions and the manufacturer's recommended settings for target species P/A surveys.

Candidate programs — those submitted for testing but not yet fully approved — are not approved for stand-alone use in P/A surveys, but may be used in conjunction with one or more approved programs.

One critical and often overlooked rule: surveyors are not permitted to switch programs from what was originally identified in their approved study plan. The software to be used must be named in the study plan submitted to the USFWS Field Office before field work begins.

When automated analysis is — and isn't — sufficient

This is the part of acoustic data analysis that is most frequently misunderstood. The MLE p-value from automated software is not a final determination of presence or absence on its own. The guidelines establish clear rules for how to interpret software output, and those rules create multiple scenarios in which qualitative analysis becomes required or strongly advisable.

If the approved software identifies one or more files as a target species and the MLE p-value is 0.05 or less, the surveyor may accept that result — presence is established at that site on that night — or may conduct qualitative analysis (i.e., manual vetting) to determine whether the result is a false positive.

If the software identifies one or more files as a target species but the MLE p-value is greater than 0.05, the guidelines direct the surveyor to either disregard the insignificant p-value and assume presence, or conduct qualitative analysis to determine if the result is a false negative.

This means that whenever a software program flags any file as a target species — regardless of the MLE p-value — the surveyor must either accept presence or conduct qualitative analysis. There is no scenario in which a low-confidence software flag can be silently dismissed without one of those two responses.

The USFWS notes that current versions of auto-ID software do not reliably provide an MLE p-value below 0.05 for rare target species when they are present in low numbers at the night level, specifically fewer than 15 files per site night. Regardless of the MLE value, if any files are identified as the target species, the surveyor must either assume presence and coordinate with the USFWS Field Office, or conduct qualitative analysis to confirm or reject presence.

The practical implication is significant: in low-activity situations — which are common for rare, federally listed species — automated software alone cannot reliably clear a site. Qualitative analysis is the mechanism that allows a surveyor to evaluate those flagged files and determine, based on expert review, whether they represent a genuine target species detection or a misclassification.

Stage two: qualitative analysis (manual vetting)

Qualitative analysis is the direct, file-by-file visual review of echolocation call recordings by a biologist with specialized expertise in bat acoustic identification. This is the "manual vetting" step, and it requires a different and more demanding skill set than deploying detectors or running automated software.

What vetting involves by software

The 2026 guidelines specify distinct vetting workflows depending on which approved software was used.

For IBAT and NLEB, when using Kaleidoscope Pro or BCID: the reviewer first examines all files labeled as IBAT and/or NLEB. If the target species is not confirmed at that step, the reviewer must then examine all other files in folders containing pulses with a characteristic frequency of 38 kHz or greater, including files in the NOID (no identification) folder.

For IBAT and NLEB, when using SonoBat: the reviewer examines files labeled as the target species in the "Spp Accp" (Species Accepted) column. If the species is not confirmed, the reviewer then examines files in the "~Spp" (Leaning Species) column — for example, MYSO, MYSO/other species, or LUSO. Files in the LUSO category could belong to either Indiana bat or little brown bat, and any files labeled LUSO should be treated as possible IBAT files.

For TCB, when using Kaleidoscope Pro or BCID: the reviewer first examines all files labeled as TCB, then reviews all other files containing pulses with characteristic frequency of 40 kHz or greater, including the NOID folder, if the target species is not confirmed in the initial review.

For TCB, when using SonoBat: the reviewer examines files in the "Spp Accp" column, then — if not confirmed — files in the "~Spp" column, such as PESU or LABO/PESU.

The standard for overturning software results

This is one of the most consequential aspects of the vetting process: it is the USFWS's preference that reviewers accept a target species auto-ID result unless the reviewer provides clear justification for why that sequence could not belong to the target species.

The burden is on the reviewer to justify a rejection — not on the software to justify its identification. A reviewer who dismisses a software flag without documented, call-level justification is not meeting the regulatory standard, and the USFWS Field Office may not accept the resulting report.

What vetting produces in the report

Qualitative analysis must include and present within a written report a comparison of the results of each acoustic ID program by site and night. The guidelines recommend providing a table that shows each species identification from the program, the suggested species ID from visual vetting, and the rationale for any changes.

If visual confirmation of the target species is achieved, presence is established and the surveyor must coordinate with the USFWS Field Office. If there is no visual confirmation of target species, no further summer or year-round active season surveys are recommended.

Who is qualified to perform qualitative analysis

This is the question that separates firms that can deliver a defensible report from those that cannot. The 2026 guidelines set specific qualification standards for the individuals who make final qualitative identification decisions.

Individuals qualified to conduct qualitative analysis of acoustic bat calls typically have: experience gathering known calls, which provides a valuable resource for understanding how bat calls change and the variation present in them; experience identifying bat calls recorded in numerous habitat types; familiarity with the species likely to be encountered within the project area; and multiple years of experience with qualitative ID skills that they have kept current. A resume or similar documentation must be submitted along with final acoustic survey reports for anyone making final qualitative identifications.

Acoustic survey reports submitted without this resume information will not be reviewed by the USFWS Field Office until the information is provided.

This is not a formality. It directly affects review timelines. Projects waiting on USFWS Field Office approval for a survey report — where tree clearing cannot proceed until that approval is received — face real schedule consequences if the report is returned for a missing resume or insufficient vetting documentation.

What the guidelines require in the final acoustic survey report

The 2026 guidelines include a detailed submission checklist for acoustic survey reports. Several items relate directly to the vetting process. Required report elements include:



• the full name and resume of any individual conducting qualitative acoustic analyses;
• the name of the service-approved and/or candidate software programs used, including versions and software settings;
• acoustic analysis software program output and summary results by site by night, including number of calls detected, species composition, MLE results, and settings files;
• discussion for any site-nights with zero bat calls;
• and, if qualitative vetting was used, a detailed analysis and results of the qualitative analysis conducted on projects where a program identified target species presence as likely, including justification for rejecting any program MLE results.

All originally recorded data — whether zero-crossing or full-spectrum — must be maintained for a period of five years and made available to the USFWS Field Office upon request. Failure to provide data when requested may result in invalidation of survey results.

Why this matters for your report

A bat acoustic survey that produces no confirmed target species detections after both automated analysis and qualified manual vetting provides defensible probable absence documentation that, under the 2026 guidelines, is valid for a minimum of five years. That is the outcome that clears your project to proceed without species-specific conservation measures — and it depends entirely on the analysis being done correctly.

A survey where automated software flagged one or two files as a target species, and a biologist without documented acoustic expertise dismissed them without documented justification, does not produce that outcome. If the USFWS Field Office requests the raw data files and finds that potentially valid target species files were rejected without adequate rationale, the survey results may be invalidated — and the project may need to repeat the survey, missing the survey window.

The vetting step is where survey quality is either demonstrated or undermined.

Volant's approach to acoustic data analysis

At Volant EcoServices, all acoustic data is manually vetted by experienced, federally permitted biologists to ensure accurate species identification. This is our standard practice on every acoustic survey we conduct — not a premium add-on. Both co-founders, Mary Gilmore and Dan Cox, hold active USFWS Section 10(a)(1)(A) Recovery Permits for Indiana bat, northern long-eared bat, and gray bat, and both bring years of qualitative acoustic identification experience to the vetting process.

We also accept third-party data vetting requests. If your firm or agency has existing bat acoustic survey data collected by another surveyor and needs qualified manual vetting and a defensible analysis report, contact us to discuss your dataset and timeline.

Volant EcoServices conducts acoustic presence/absence surveys, full acoustic data analysis and qualitative vetting, and USFWS study plan coordination for state and federally listed bat species across the eastern United States. Learn about our bat survey services or contact us about your project or dataset.