Proposal talk:Tag:natural=fungus

Totally inappropriate to map such things on OSM: too small, too many, too ephemeral, not readily verifiable by other mappers, GPS resolution not adequate. (as SK53 said on the natural=flowers page). --Polarbear w (talk) 22:51, 26 July 2017 (UTC)

This is worth mapping

I think people commonly mistake fungus for mushroom - and the original proposal did the same mistake. No, they are not small insignificant and short lived. Instead they are some of the biggest, heaviest and oldest organism living on earth: https://en.wikipedia.org/wiki/Honey_fungus

It would be a joke not to map these if we map trees! RicoZ (talk) 20:21, 2 February 2019 (UTC)

@RicoZ, they are just too ephemeral. Whereas street trees may last 100 years or so (often longer than the buildings), a fungal fruiting body will last days. Whereas we know how many trees there are in a country like GB (around 80 million Ash trees) and in most cities (typically low 10,000s for a city of around 500k), the number of ephemeral fungal fruiting bodies will be much higher. Also there are perfectly good applications - Observation.org; iNaturalist, iRecord - for recording fungi, all of which use OSM as a base layer. Lastly, I am interested in so-called microfungi which often grow on particular leaves of particular plants. Hope you're not adovating mapping those. SK53 (talk) 15:05, 3 February 2019 (UTC)

This is not about every single mushroom you will find in the forest or elsewhere but about the fungus that grows underground and may grow many thousands of years old. There are only a few of them that have been discovered and documented like https://www.biorxiv.org/content/10.1101/377234v2 but those are certainly more long lasting and bigger than most of the trees.

Thanks for mentioning the microfungi, I would not micro-map them but wondering if the area and pattern of their prevalence should be mapped? I am thinking about vegetation levels so might somehow include the idea of "parasitic" vegetation into my thoughts. RicoZ (talk) 21:31, 4 February 2019 (UTC)

I agree that mapping places in a forest where you can usually gather edible fungi (and other edible seasonal plants) can be useful to many. Despite the fact that you can not see them all year round, the bulk of their root systems cover a well defined part on the ground, and thus defines the expected area that you can harvest. Mapping each cluster on its own sounds excessive and hard to maintain, but probabilistic measures of areas with highest concentration sounds reasonable. According to the Wikipedia definition: "A mushroom, or toadstool, is the fleshy, spore-bearing fruiting body of a fungus, typically produced above ground on soil or on its food source." Thus, it would be excessive to map each head of mushroom, but it is much more reasonable to map the extent of the fungi root system that produces these mushrooms, so natural=mushroom would not make sense. See also taxon=* Bkil (talk) 08:42, 14 April 2019 (UTC)

For the "probabilistic mapping" I am considering something different and more general, would prefer to keep natural=fungus for well known individual fungi. RicoZ (talk) 21:55, 14 April 2019 (UTC)

So, would you care to share your proposal then? As size of a fungus is in the micrometer range, I doubt that you would want to map each of them individually. Rather, by "probabilistic mapping" I meant to mark the area that visibly encompasses the majority of the fungal system with high probability (as it is infeasible to tell by micrometer accuracy). This may be as small as the vicinity of a tree or two, or it may spam hundreds of meters along a given path. Mapping is just as useful if I can only state that 50% of the area along the right side of this path is covered with this fungus/garlic/berries/etc. My reasoning above states that the mapping detail should be sufficient to support mushroom gathering (i.e., know where to look for them inside a large forest), and to be easy to maintain. Bkil (talk) 16:34, 16 April 2019 (UTC)

How to map

How could a mapper identify such an area for mapping, or how can a mapper verify that an already mapped area is indeed covered by such a fungus? --Lyx (talk) 10:54, 3 February 2019 (UTC)

Good question. We have to rely on the scientists publishing their data (https://www.biorxiv.org/content/10.1101/377234v2) and so far they published gps tables of points where a particular individual has been found (https://www.biorxiv.org/highwire/filestream/130618/field_highwire_adjunct_files/1/377234-2.xlsx). There is also a sketch of a map inside the paper. Not sure if the current proposal to map them as area is flexible enough if all we have is a collection of known points. It might be possible to extrapolate an area from the points - we are not mapping them in 3D anyway so a good approximation may be sufficient? Or should we introduce collections of points? RicoZ (talk) 21:31, 4 February 2019 (UTC)

After some thought I would definitely prefer to do the abstraction/generalization and map an approximate area instead of a collection of "precisely known" points. The generalization is much more generally useful, the error margin should be fairly small and it is possible the margins will be somewhat changing over time. RicoZ (talk) 22:10, 9 February 2019 (UTC)