State of India’s Birds 2023: A framework to leverage semi-structured citizen science for bird conservation
Bibliographic record
Abstract
Abstract Birds and their habitats are threatened with extinction around the world. Regional assessments of the ‘State of Birds’ are a vital means to prioritize data-driven conservation action by informing national and global policy. Such evaluations have traditionally relied on data derived from extensive, long-term, systematic surveys that require significant resources, limiting their feasibility to a few regions in the world. In the absence of such ‘structured’ long-term datasets, ‘semi-structured’ datasets have recently emerged as a promising alternative in other regions around the world. Semi-structured data are generated and uploaded by birdwatchers to citizen science platforms like eBird. Such data contain inherent biases because birdwatchers are not required to adhere to a fixed protocol. An evaluation of the status of birds from semi-structured data is therefore a difficult task that requires careful curation of data and the use of robust statistical methods to reduce errors and biases. In this paper, we present a methodology that was developed for this purpose, and was applied to produce the comprehensive State of India’s Birds (SoIB) 2023 report. SoIB 2023 assessed the status of 942 bird species in India by evaluating each species based on three metrics: 1) long-term change: change in abundance between the year 2022 and the year-interval pre-2000; 2) current annual trend: mean annual change in abundance from 2015 to 2022; and 3) distribution range size. We found evidence that 204 species have declined in the long term, and 142 species are currently declining. We present and discuss important insights about India’s birds that can guide research and conservation action in the region. We hope that the detailed methodology described here can act as a blueprint to produce State of Birds assessments from semi-structured citizen science datasets and springboard conservation action in many other regions where structured data is lacking but strong communities of birders exist. Open Research Statement The primary data are already publicly available on eBird (Sullivan et al. 2014), and other data are already published in a GitHub repository (stateofindiasbirds 2024).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.023 | 0.028 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.010 | 0.008 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.007 | 0.005 |
| Open science | 0.003 | 0.009 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".