EcoSpold 1 ยท Calculation comparison

BAFU 2025 v2, Water use

A comparison of published impact results with VoLCA calculations from the same source data and the same method, in the implementation named below. Points on the diagonal agree.

Engine version
volca 0.18.0 (85ea97a, v0.18.0, linux)
Measured on
Comparison measurements and reference
DatabaseBAFU 2025 v2
Source formatEcoSpold 1
Impact method collectionEF 3.1
Method implementationopenLCA 2.8.2
Impact categoryWater use
Unitm3 world eq
Activities compared11,536
Median ratio (VoLCA / published)0.6239
Comparison band1%
Activities outside the band11,348
Share outside the band98.37%
VerdictDisagrees, traced to the publisher's own results
Known reasonBAFU's water column does not follow the inventories shipped with it. Net water is a remainder of withdrawals and returns about a thousand times larger, so slight inventory changes move it: computed on the shipped inventories it rises by a factor 1.68 from this release to 2026 v1, while BAFU's column stays put (median ratio 0.996). VoLCA on the 2026 v1 inventories lands at 1.016 of this column. 653 rows also repeat their climate change value, shifted by a power of ten.
Points drawn2,082
Largest relative deviation-100%
Activity with largest deviationOperation, passenger car, natural gas {CH}
Published referenceBAFU-2025 version2_LCIA Results.xlsx
BAFU 2025 v2, Water use
Each point is one activity: what BAFU 2025 v2 published for Water use (m3 world eq), against what VoLCA computes from the same EcoSpold 1 source under the EF 3.1 factors of openLCA 2.8.2. The black line is equality and the grey band is 1%. 11,536 activities were compared; 11,348 fall outside the band, more than a picture can hold, so 1,923 of the 11,536 are drawn, evenly spaced over the published value whichever side of the band they fall, and the 159 of the 200 farthest from equality that this walk missed are added. Where several fall on one spot, at most three markers are drawn there, since a fourth would not darken it. Median ratio 0.6239, 98.37% outside the band. This misses the agreement a release is held to (at most 5% of activities outside the band and a median ratio inside it), for a known reason: BAFU's water column does not follow the inventories shipped with it. Net water is a remainder of withdrawals and returns about a thousand times larger, so slight inventory changes move it: computed on the shipped inventories it rises by a factor 1.68 from this release to 2026 v1, while BAFU's column stays put (median ratio 0.996). VoLCA on the 2026 v1 inventories lands at 1.016 of this column. 653 rows also repeat their climate change value, shifted by a power of ten.