iNaturalist vs eBird

[ram_k]

@pmeisenheimer
Very well said – and as of now an important on the ground difference.

Ebird “allows” sharing of observations - acknowledging that people do observe together and it emphasizes what you say

[hdemann]

iNaturalist is much better than ebird

1. The peer review is open to more people and transparent. On eBird, your data can be rejected by a single local reviewer, and you might never even know that it happened. I have seen a lot of records on eBird that are wrong because of the mistakes of one reviewer.
2. iNaturalist has API available, eBird does not. The result is that you are punished for not using eBird as your primary database. EDIT: ebird does have an api that can be used to view overall user data. My comment about the API is that you cannot use it to manage your own personal observations.
3. The entire “checklist” process on eBird is not scientific as it influences a user toward certain identifications.
4. eBird data is skewed by listers who submit data about rare bird sightings. One rare bird in a location could be shown as common due to multiple lists even if it is a vagrant.
5. The “hotspot” concept on eBird is not consistent as some hotspots are very local while others are broad.
6. As mentioned, eBird ignores non-bird data.

For these reasons, I think that iNaturalist is clearly the superior platform.

[raymie]

For the purposes of people interested in birds, eBird is the superior platform due to its checklist format and ease of locating rarities. eBird does have an API available, by the way.

[swampster]

@hdemann

iNaturalist is much better than ebird

Both are extremely useful platforms with very different data (and use) objectives. I use both in a variety of ways for my research. eBird is the far superior platform for abundance and distribution data, and by extension, habitat use, migration flyways, etc. And iNat is better for looking at interactions, sources of mortality, and to a lesser extent, behavior.

@hdemann

On eBird, your data can be rejected by a single local reviewer, and you might never even know that it happened.

While you do have to actively check, you can use the “Species Maps” to check records of interest to know if they have been rejected/accepted. I have never had a reviewer reject an observation where I provided the appropriate documentation according to eBird’s guidelines, and I have lived/traveled to a variety of places with different reviewers. That’s not to say “bad” reviewers don’t exist, but they aren’t commonplace.

@hdemann

iNaturalist has API available, eBird does not.

This is not true. eBird does have an available API.

@hdemann

The entire “checklist” process on eBird is not scientific as it influences a user toward certain identifications.

I think you are referring to how eBird provides a list of species likely in your area from which you build your checklist. This is no different from iNat’s computer vision which provides suggestions based on what’s “expected nearby”. On both platforms, you can add IDs for any species globally on any checklist/observation.

Building a checklist from likely species is useful for two reasons. Ease of use: it’s way easier for me to build my checklist if I’m not having to scroll through >11,000 global bird species. Data quality: how often do we see iNat observations where someone has added an incorrect ID because they selected a computer vision suggestion prior to inputting the observation location. On eBird, you can change your settings and use the complete global list, but I’ve seen people do this and then submit Scissor-tailed Kites in Florida that were undoubtedly Swallow-tailed Kites.

@hdemann

eBird data is skewed by listers who submit data about rare bird sightings. One rare bird in a location could be shown as common due to multiple lists even if it is a vagrant.

This is actually far more prevalent on iNat than it is on eBird. eBird encourages complete checklists, meaning the common species get documented along with the rare. This is not the case for iNat. I eBird daily, and have far more checklists with only common species than ones that include rarities. My iNat observations skew heavily towards things that are new or unusual to me.

@hdemann

The “hotspot” concept on eBird is not consistent as some hotspots are very local while others are broad.

While eBird hotspots aren’t perfect, they should, per the guidelines, have easily defined boundaries and are spatially explicit. This is no different than the wide range in accuracy circles you find on iNat (and actually, when I submit from the app, it often does not include an accuracy circle at all for the observation). Additionally, many researchers using eBird data use the eBird tracks created by those submitting from the app (most common submission method), not the hotspot coordinates.