World Migratory Bird Day 2026: How Citizen Observations Inform Business Nature Risk

Volunteer bird counts have helped establish 956 Ramsar Wetlands of International Importance covering 1.5 million square kilometres, identify 2,701 Important Bird Areas covering 1.75 million square kilometres, and underpin 2,721 EU Special Protection Areas covering 418,000 square kilometres.

Those are designations, not awareness campaigns. An EU Special Protection Area carries a legal consequence for any project that might affect it. So while corporate teams debate whether amateur sightings are reliable enough for site screening, that same body of observation has already become the constraint map their permits are assessed against.

The question is not whether to trust the data, but which of the two very different things it does you are relying on.

The Day, And The 2026 Theme

World Migratory Bird Day is observed twice a year, on the second Saturday of May and of October, because migration peaks at different times in the northern and southern hemispheres. The 2026 dates are 9 May and 10 October, which falls this Saturday.

The theme is "Every Bird Counts: Your Observations Matter!", which is unusually on point for a corporate audience. The campaign is run by the CMS and AEWA Secretariats with the East Asian-Australasian Flyway Partnership and Environment for the Americas, and is built this year around community science: that volunteer observations help track migration routes, population trends and habitat change.

It also marks the sixtieth edition of the International Waterbird Census, which has run since 1967, covers 189 countries and territories and surveys some 67,000 wetlands. The IWC is where the designation figures above come from.

The Designation Channel Is Where Volunteer Data Already Binds You

The mechanism is direct.

Ramsar Criterion 5 qualifies a wetland if it regularly supports 20,000 or more waterbirds. Criterion 6 qualifies it if it regularly supports 1 per cent of the individuals in a population of one species or subspecies of waterbird. Both are counting criteria, satisfied in practice by the coordinated volunteer census the IWC organises. Key Biodiversity Area criteria work on comparable logic, with thresholds set as percentages of global population, and Criterion D1 covering demographic aggregations of highly mobile species at stopover and bottleneck sites.

Once a site is designated, the evidentiary burden flips. Under Article 6(3) of the Habitats Directive, a plan or project likely to have a significant effect on a protected site requires an appropriate assessment, and the Court of Justice held in Case C-258/11 Sweetman, decided on 11 April 2013, that such an assessment cannot have lacunae, must contain complete, precise and definitive findings and conclusions, and must be capable of removing all reasonable scientific doubt about the effects on the site.

The asymmetry is the point. A volunteer count is sufficient to create the designation and nowhere near sufficient to discharge the assessment that designation triggers. Companies that notice only the second half tend to dismiss citizen data as unreliable, which misreads what it is doing at each end.

What A Sighting Record Actually Is

eBird passed two billion observations in June 2025, across 150 million checklists from 1.1 million eBirders. The volume invites a category error, so be precise about what sits inside a record.

A casual sighting is presence-only data. It tells you a species was seen somewhere at some time by someone. It does not tell you how long they looked, how far they walked, whether they recorded everything they saw, or how many other observers never visited at all.

Those gaps have names. Johnston and colleagues, writing in Diversity and Distributions in 2021, set out the main biases in community science data: spatial bias, because volunteers visit places near home, accessible, species-rich or already protected; variable effort, because duration, distance, time of day, weather and observer count all change detection probability; selective reporting; and imperfect detection, which turns a present species into a recorded absence.

Their results are the part worth carrying into a procurement conversation. Models built on presence-only or incomplete-checklist data performed substantially worse than better-specified models and, more damningly, their estimates were poorly correlated with those of the best model. The biggest improvement for one study species came from using complete checklists, where the observer confirms they reported everything detected, so non-detection carries information. Spatial subsampling, the correction most people reach for first, barely changed performance.

Two datasets therefore hide inside the same platform: a pile of sightings, and a set of effort-annotated complete checklists that supports inference if filtered properly. Treating them as interchangeable is where corporate screening goes wrong.

Where Local Observations Genuinely Help

Two uses hold up.

Hypothesis generation and timing before you scope a survey. If records near your site show a species of concern that nobody mentioned in the project brief, you know to commission the right survey. Records also tell you when passage actually occurs at your latitude, which decides when a survey is worth paying for, because a wintering bird survey run in the wrong months produces an expensive null result.

Challenging a consultant's baseline. If the ecology chapter reports no records of a species that appears repeatedly in nearby public datasets, that is a question worth asking before a regulator asks it. It is the most useful application here and almost nobody does it.

Strategic siting tools already work this way. The Avian Sensitivity Tool for Energy Planning, built by BirdLife International with the Asian Development Bank, scores 5 by 5 kilometre squares for sensitivity to wind, solar and power lines across nine countries, from Egypt and Kenya to Uzbekistan, India and Australia. BirdLife says the Australian maps drew partly on data from thousands of BirdLife Australia citizen scientists working alongside professional ornithologists. It is a layer for deciding where to look, not a substitute for project-level assessment.

What They Cannot Prove

Absence. No records is almost never evidence of no species. It is usually evidence of no observers, and the spatial bias above means the two get confused in exactly the places industrial sites tend to be.

Abundance at your site. Unadjusted counts conflate how many birds are present with how hard people looked. A location with a bird club nearby will out-record an identical location without one.

Threshold compliance. Whether a site regularly supports 1 per cent of a population needs standardised, repeated counts with known methodology. The KBA Global Standard is candid on the split: locality data may legitimately come from literature, museum records, GBIF and citizen science platforms, but confirming presence and reproductive units points to recent data, local knowledge-holders or new field survey.

Causation. Records cannot tell you whether your operation caused a change; that needs a before-and-after design with a control.

And an Article 6(3) assessment. Findings capable of removing all reasonable scientific doubt is not a standard presence-only records reach, and a consultant who submits one built mainly on public sightings has handed any objector a straightforward ground of challenge.

The Radius You Are Now Obliged To Choose

This lands in reporting as well as permitting. Simplification dropped the ESRS E4 requirement to disclose every site and break sites down by ownership type, which sounds like relief until you read what replaced it. E4-5 asks for the locations in your own operations related to material impacts, risks or opportunities, then the biodiversity-sensitive areas linked to material negative impacts by name and type, and the activities causing them. E4-2 asks for your policies on sites in or near such an area.

The ESRS definition has two families: areas protected by legal or other effective means, including Natura 2000, UNESCO Natural World Heritage Sites, Ramsar sites and nationally protected areas; and areas scientifically recognised for biodiversity importance, including Key Biodiversity Areas, Ecologically or Biologically Significant Marine Areas and habitats of IUCN Red List species. The second family carries no permit and appears in no consent you hold.

The detail that catches people is that "near" has no fixed distance. Whether a site outside a sensitive area counts as near one is settled by defining that site's area of influence, using buffer distances specific to the type of activity and drawn from regulatory requirements, science-based recommendations and industry best practice. So a radius picked without ecological basis is a number an assurance provider can ask you to justify and you cannot, and 1 kilometre that works for a warehouse will not work for a plant discharging into a river that feeds a designated wetland twelve kilometres downstream. Observation data will not set that radius, but it will tell you whether species using that wetland are also recorded on your own land.

India: Thirty Million Records, Still Not A Site Baseline

India illustrates both halves, because the dataset is world-class and still does not answer a site-level question.

State of India's Birds 2023 assessed 942 species using more than 30 million observations from over 30,000 birdwatchers, flagged 178 as high conservation priority, and found 60 per cent of species with long-term trends in decline. That is a serious national evidence base, produced largely by volunteers, and it is now routinely cited in policy.

The project's own framing is the instructive part. It notes that systematic monitoring using structured methods is what provides the site-specific information needed to explain local population change. The people who built the largest citizen bird dataset in the region are telling you it does not substitute for structured monitoring at a site.

For an Indian environmental clearance the implication is narrow. Public records are excellent for working out which species your terms of reference should cover and which seasons your surveys must span. A baseline assembled from downloads rather than seasonal fieldwork is the kind of weakness that surfaces during appraisal rather than before it.

The Index Nobody Funded

One more reason this is a corporate matter, not a conservation one.

The Nature Restoration Regulation, Regulation (EU) 2024/1991, in force since 18 August 2024, requires restoration measures across at least 30 per cent by 2030 of the area of listed habitat types not in good condition, rising to 60 per cent by 2040 and 90 per cent by 2050. Mind the base, which secondary coverage garbles routinely: it is degraded area put under restoration, not habitat in good condition.

Birds sit apart from the three agricultural indicators Member States choose two of, none of which is a bird. The bird obligation is its own: an increasing trend in the common farmland and forest bird indices by 2030 and beyond, under the Annex V and VI methodologies, applied nationally against species sets relevant to each country.

Draft national restoration plans were due by 1 September 2026, five weeks ago. Measures landing on farmland reach agricultural supply chains, so companies sourcing food in Europe should read those plans now, not when a supplier mentions a land use change.

The index is produced by the Pan-European Common Bird Monitoring Scheme, coordinated from the Czech Society for Ornithology, out of annual breeding bird surveys run by volunteers across 26 EU countries. The scheme's 2023 reporting described supplying the Commission with the index targets and the cost of national monitoring, while noting its own routine activities were then unfunded. Whether that has since been fixed is not something to assume either way, but an EU policy indicator produced by a volunteer network on project funding is worth understanding before you depend on it.

A Sequence That Survives Review

Start with designations, not sightings. Run your sites against protected area and KBA layers using IBAT or equivalent, and record the negatives too, because the absence of a designation within your buffer is a disclosure you may have to evidence.

Scope surveys from what the records suggest, by season, then diarise their expiry. Published guidance points to roughly four to six monthly visits for wintering and non-breeding birds and two to three for breeding birds, so a credible baseline takes a year rather than a quarter. CIEEM's April 2019 advice note then sets the shelf life: under 12 months, likely valid; 12 to 18 months, usually still valid unless mobile species may have moved in; from 18 months to three years, a professional ecologist must visit the site and state what remains valid; beyond three years, most or all surveys need redoing. Put the 18 month boundary in the project plan, because that is where the cost stops being a desk exercise.

Citizen observation is strongest at the scale where it was collected and weakest at the scale where decisions get made. It will tell you a flyway runs through your region, which is more than most risk registers know. It will not tell you what is on your site this season, and on the day a regulator or an assurance provider asks, thirty million records in a public database will not answer for you.

Position as of 8 October 2026. Protected area designations, Key Biodiversity Area boundaries, AVISTEP country coverage and the status of national restoration plans all change. Survey requirements and validity periods differ substantially by jurisdiction and consenting regime, and the CIEEM bands cited are professional guidance rather than a legal rule. The ESRS E4 wording follows EFRAG's technical advice of 30 November 2025 on the simplified standards; the operative text is the adopted delegated act in the Official Journal and should be read there. Confirm designation status against the relevant competent authority and current datasets, and take professional ecological and legal advice on any specific site, assessment or consent.

Sources

World Migratory Bird Day, Wetlands International, Ramsar Convention on Wetlands, Key Biodiversity Areas Partnership, Johnston and others, Cornell Lab of Ornithology, Global Biodiversity Information Facility

This article is intended for general professional information and does not constitute legal, financial, or investment advice.