A city-region planning team is drafting its green-infrastructure baseline and wants the headline shares — tree cover, built-up, grassland, water — for Greater London from the ESA WorldCover map rather than from a survey nobody has budget to repeat. The figures go in a document that will be quoted for five years, so they have to come from the raster as it is, with the method stated.
From the land-cover raster, report the share of valid pixels (class ≠ 0) in each of these classes, as a percentage to two decimals:
tree_cover_pct — class 10built_up_pct — class 50grassland_pct — class 30water_pct — class 80and classes_present — how many distinct non-zero classes occur in the clip.
Pixel shares, not areas: the grid is geographic, so a pixel in the north of the clip is slightly smaller on the ground than one in the south. Over Greater London the difference is under one percent, and the baseline states 'share of pixels' for exactly that reason.
These are where the teaching is. Read them twice.
Class shares from a land-cover product are the opening figures of every urban green-infrastructure baseline, and the method sentence beside them is what lets the next team reproduce them.
Work the problem in whatever tool you like, then enter the answers here.
tree_cover_pctbuilt_up_pctgrassland_pctwater_pctclasses_present5 scored, 0 informational
Tree-cover share
30%Built-up share
30%Grassland share
20%Water share
10%Distinct classes present
10%