Like Dave I have come to the view that mapping individual fields as farmland is a good way to do it.
I use farmland=arable & farmland=pasture. This still does not cover cases of permanent grassland which are not used for pasture. I can see the value of farmland=livestock for things like pig rearing which I have never tried to map (largely because I think they move around). The advantage of doing it field by field is that one can use real local knowledge from surveys, can discriminate between fields which are apparently similar on aerial imagery, can add more detailed tagging (for instance I have used some plant community tagging in one or two areas, or one can mark pastures with ridge and furrow). The absence of large polygons is of course a significant benefit. Meadow as a synonym for any old bit or rural grassland creates huge problems if we ever want to identify real meadows which are one of the scarcest habitats in Britain now. Such usage also covers a range of quite different things. It is well established that people no longer know what a true meadow looks like, but I'd hope we can be more sophisticated. Most meadows will be available in various classes of Natural England habitat open data. CRW have similar datasets, as do SNH. Unfortunately nothing is available for Northern Ireland, although I've been informed by a former head of the NIEA that Perennial Ryegrass is now the national plant : in other words most agricultural grassland is heavily improved whether as pasture or as leys for silage crops. W.r.t. Mark's point, there are open data <https://data.gov.uk/dataset/fb19d34f-59e6-48e7-820a-fe5fda3019e5/crop-map-of-england-crome-2018> from RPA for the past few years on agricultural usage, so it is possible to use a bit more than guesswork. Note also much aerial imagery in the countryside may be significantly out-of-date. I've both walked through areas which look like arable on aerials with sheep on them and areas of pasture have subsequently been ploughed (notably on what was formerly part of Muston Meadows NNR). The RPA data at least allows a more up-to-date picture. It's a hexgrid derived from remote sensing, but in most cases can be interpreted. Regards, Jerry On Sat, 14 Dec 2019 at 16:24, Edward Catmur via Talk-GB < talk-gb@openstreetmap.org> wrote: > Some mappers use meadow for permanent pasture, on the basis that this is a > fundamentally different use of land to putting it under the plough. > > Others believe that meadow should be reserved for "real" meadow, and that > permanent pasture should be distinguished from cropland by some combination > of sub tags. > > On Sat, 14 Dec 2019, 16:09 Martin Wynne, <mar...@templot.com> wrote: > >> > I would say yes, as I believe both arable & livestock is farmland. >> >> Thanks Dave. >> >> But in that case, how on OSM do we differentiate between the two? >> >> It seems silly that in some areas of OSM we can go into ridiculous >> detail, such as whether a bench seat has a backrest, but vast tracts of >> land which visually look very different are classed as one and the same? >> >> cheers, >> >> Martin. >> >> _______________________________________________ >> Talk-GB mailing list >> Talk-GB@openstreetmap.org >> https://lists.openstreetmap.org/listinfo/talk-gb >> > _______________________________________________ > Talk-GB mailing list > Talk-GB@openstreetmap.org > https://lists.openstreetmap.org/listinfo/talk-gb >
_______________________________________________ Talk-GB mailing list Talk-GB@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-gb