Journal archives for May 2022

May 16, 2022

iNat Staff Mt Tam Hike

Had a nice hike with a few iNat staff members (@abhasm, @tiwane, @kueda, and @albullington) last Friday on Mt. Tam
For me the highlight was the group of Fairy Moths swarming around their Ocean Spray host plant

Posted on May 16, 2022 05:54 PM by loarie loarie | 0 comments | Leave a comment

May 24, 2022

iNaturalist is hiring two developers!

We’re thrilled to announce that iNaturalist is hiring two developers to join our team to help us improve our mobile apps and continue to scale our infrastructure. You can learn more about these positions in our new jobs page.

We’re looking for full-time staff who are interested in careers with iNaturalist. Candidates may be remote and need not be in the United States.

Thanks for helping us with these searches by sharing these openings with your networks!



A few iNaturalist staff (@loarie, @abhasm, @albullington, @tiwane, & @kueda) exploring Mount Tamalpais in May

Posted on May 24, 2022 06:55 PM by loarie loarie | 12 comments | Leave a comment

May 27, 2022

We’ve passed 100,000,000 verifiable observations on iNaturalist!

If you made 1,000 observations a day, every day, it would take you 274 years to generate 100 million observations. This milestone shows what people can do by working together. The iNaturalist dataset is something we’ve all made together, but it’s larger than any one of us. We hope everyone is as proud of this accomplishment as we are. Together, the iNaturalist community has created a unique window into life on Earth and hundreds of thousands of species with whom we share the planet. Thank you!

We know that even more potential for iNaturalist lies ahead. To fulfill our mission of connecting people to nature and advancing science and conservation, we’re working on a strategy to reach 100 million naturalists by 2030. This requires investing in technology improvements, so we’re now searching for two new software engineers to join the iNat team. Please spread the word to help us find great candidates.

100M observations across time

The graph below is a slightly different perspective of how these 100M observations have accumulated over time. The spiral begins in early 2015 when we reached our first million. Each band on the spiral represents another 1M observations (here's links to earlier posts about the 5M, 6M, 25M and 50M milestones). Each revolution represents a year. You can see the major patterns in the graph: the number of observations growing over time and the bands becoming closer together, the seasonal cycle where the spiral is thinner in the northern hemisphere winter, and the City Nature Challenge annual bioblitz bulge in April.

100M observations across species

Just as these observations aren’t evenly distributed across time, they are not evenly distributed across species. If we sort observations by species in descending order, we can count how many species it takes to cross 1M observations. A group with the top 5 most observed species cross the 1M threshold alone (Group 1), as does a group with the next 8 species (Group 2).

The graph below continues descending with groups of species needed to cross 1M observations through all 372,327 species. In other words, each colored square represents a group species and ~1M observations. Species number 372,327 is reached in the 75th group and the remaining ~25M observations are identified at coarser ranks than species (genus etc.).


Note that with each descending group more and more species are needed to cross 1M observations. For example, group 50 consists of 378 species each represented by 724 to 836 observations. Group 72 consists of 10,854 species represented by 77 to 111 observations. Because our computer vision model includes species for which we have around 100 observations, this is the point where species become too rare to be included in the model yet. The last group consists of 65% (243,544) of all the species observed on iNaturalist, each represented by only a few (1-15) observations. In other words, the top 5 species on iNaturalist have about as many observations as the bottom 243,544 species.

100M observations across space

These observations are also not distributed evenly in space. In some places, such as the area around Austin, Texas, the density of observations is high with about 88 observations per square kilometer. This means that a 60 km radius circle with an area of 11,310 km2 contains 1M observations.

In other parts of the world, the density is much lower, meaning that circles encompassing 1M observations are much larger. The map shows 100 example circles each of which contains 1M observations. All together, these circles encompass 100M observations. Many circles across much of North America, Europe, and places like Taiwan and Cape Town are small, reflecting high densities like those found near Austin, Texas. But in other places like west Africa, western Australia, and eastern Russia the circles are very large, reflecting the low density of observations (as low as 0.02 observations per km2 in eastern Russia). Note: The circles aren’t projected perfectly in the figure so they might not exactly encompass 1M observations as drawn.

While every observation on iNaturalist represents a meaningful contribution, these seasons, species, and parts of the world with underrepresented numbers of observations are opportunities for the iNaturalist community to make outsized contributions towards bringing the natural world into focus through the iNaturalist dataset.

If you’re inspired to donate in celebration of the 100 million observation milestone, you can donate in dozens of currencies to support iNaturalist. iNaturalist is a not-for-profit joint initiative of the California Academy of Sciences and the National Geographic Society, and all donations are received by the California Academy of Sciences (Tax ID: 94-1156258).


Donate to iNaturalist



Thank you to everyone who has contributed to this milestone! We look forward to the next 100 million observations.

Posted on May 27, 2022 10:27 PM by loarie loarie | 24 comments | Leave a comment