Located aquifer inventory

Generated by inventory_aquifers.py. Every known aquifer body within 40 km of the site-screen box, with its location. No distance to the proposed store appears anywhere in this file.

Why the box and not the store

The store at E330255 N122245 is one candidate location for one component of a scheme that has not been designed. Every previous aquifer table measured from it, and the standing 625k table cuts at 25 km from it - so a design assumption was deciding which geology got into the evidence base at all. The Bridport Sand missed that table by 1.4 km of an assumed radius from an assumed point.

Distances here are measured from dem_bounding_box = E315,000.0-360,000.0 N115,000.0-145,000.0, the ground where an intake or reservoir could actually go. A body that overlaps the box is at 0 km. The box is a data boundary; it does not move when the design does.

This inventory does not rank anything. It locates. Once a real intake or reservoir site exists, call bodies_near(inv, E, N, km) - or python inventory_aquifers.py somerset --site E,N - to assess each body for each role against THAT site.

What is in and what is out

• 335 bodies, 7,887 km2 within 40 km of the box
• Class 3 excluded: 263 polygons. ‘Essentially no groundwater’, and it records no flow mechanism - not an aquifer, so not in an aquifer inventory
• bodies below 0.25 km2 dropped: 126
• polygons are dissolved by unit then split into CONNECTED bodies, so a row is something water could move within - the same compartment rule aquifer_registry.yml works to

COVERAGE GAP, STATED NOT ASSUMED. 2,001 km2 of the 12,368 km2 search halo (16%) falls outside the bounding_box the 625k was downloaded to, so any body there is invisible to this table however close it is to the site box. Widen bounding_box and re-download to close it. This is the same class of artefact project.yml already records for the borehole and LiDAR boxes: a data box deciding the answer.

By band and flow mechanism

Intergranular is broken out because it is the only column that decides whether a body can STORE water rather than merely transmit it.

Every body

role is carried from screen_portfolio.py via our own 50k model. A blank our_formation means the 50k screenshot does not reach that body - we have not looked, which is not the same as nothing being there.

nearest to box is the point ON the body closest to the site box - the location to use. For a big body the centroid is meaningless: the White Chalk body spans 71 x 79 km, so its centroid lies about 40 km from the ground where it actually comes near us. Centroid and bounding box are in the CSV for completeness, not for siting.

extent is the longest dimension. Read it against the compartment rule in aquifer_registry.yml: water moves about 16 m through this rock in the 229-day surplus-to-deficit lag, so ANY body here longer than a few km holds several independent stores. No row in this table is one tank, and none of these areas may be multiplied into a capacity.

role is the role of the body’s DOMINANT 50k formation, and #fm counts how many distinct 50k formations we map inside it. Where #fm is greater than 1 the role describes part of the body only - this project’s standing finding is that one 625k polygon can hold 6-9 distinct 50k units, and at least one Class 1A body below is dominated by SEAL. Class and role disagreeing is the mosaic showing through, not an error.

The roles these will be assessed against

A role is a claim about what a body could DO for a scheme. It cannot be settled without knowing where the water is taken from and where it is put - which is exactly why this file stops at location.

← All notes · More from Aquifer Storage