Rather puzzled by the depiction of agricultural land in large areas of France. Much of this land has great blobs of light brown (pale green on my Android) these ‘blobs’ bear no relation whatsoever as to how this land is farmed or managed.
In England, certainly in the part I am familiar with, quite rightly, farmland has no depiction as to its usage i.e. pasture or arable. In the majority of cases these can be interchangable anyway. Such is even more common in many parts of France. One season a field might be ploughed and tilled, then the next year it could be in pasture.
Landuse data for France and many other countries comes from CORINE (usually 2006) a EU wide landuse/landcover project based on lowish resolution satellite imagery and roughly with a minimum parcel size of 25 ha. It was not necessarily ground truthed: but probably works fine for broad-brush statistical purposes as the inaccuracies will not tend to be systematic.
In general it is pretty inaccurate from an OSM point of view. I correct (improve) the polygons when I come across them, but not much more than that.
In some places it is very obviously wrong: I have a slide of a Pine woods and a Chestnut ‘orchard’ at way above the grape growing altititude which came from CORINE as vineyard.
Also unfortunately OSM tagging of agricultural land tends to be driven by a desire to make the map green: large swathes of various countries are tagged landuse=grass which is totally uninformative.
Many thanks indeed for your very informative response, I do realise what you are saying, so looks like there is not a great deal we can do about this. Even so, I’m pleased to know how this came about. Thanks again, Philip.