Chiwara Team — Quantities Module Attribution Requests¶

Status: Pending team response. This document can be sent to the Chiwara field engineering team as-is.

Context: The quantities module reverse-engineering (see quantities-spec.md and quantities-formula-reconciliation.md) identified a set of parameters where the current reference workbook either (a) omits a needed value, (b) uses a proxy that is methodologically incorrect, or (c) references an unidentified source. The tool cannot be made defensible under ISO 14071 critical review until these are resolved.

Each item below is classified: - BLOCKED — the module raises ValueError at runtime until this is provided. No default, no placeholder. A BLOCKED parameter means the calculation simply cannot run. - EXPERT-JUDGEMENT — the module can run with the proxy value from the workbook, but the result is flagged as EXPERT-JUDGEMENT in the audit trail, not SOURCED, and will be called out in an ISO 14071 review.

Item 1 — BLOCKED: Slab mesh specification (row 51)¶

Module location: quantities, BoQ item B.II.1.b — "Reinforced concrete slab, 300 kg/m³, 10 cm thickness".

The problem: The current workbook has no rebar (Steel Rebars, column M) for this slab. Every other reinforced concrete item in the workbook (isolated footings, columns, beams — rows 9, 11, 12) includes rebar at 100 kg/m³. A 100 mm structural RC slab without any reinforcement is not structurally plausible. The Python module cannot include a rebar formula for this slab without the correct mesh specification — and it cannot safely apply the 100 kg/m³ density from the 350 kg/m³ structural recipe to a 300 kg/m³ lightly-loaded floor slab.

What we need from you: 1. The reinforcement specification for this slab: mesh type (e.g. ST30, ST40, treillis soudé), bar diameter, spacing, or equivalent density in kg/m³ or kg/m². 2. Who provided the specification (name, date, project reference). 3. Whether this specification is drawn from a standard (NF, EN, or local) or from a project-specific structural design.

Why this is BLOCKED: If we invent a rebar density, the module could understate or overstate total project steel by ~163% (see quantities-formula-reconciliation.md §4.1). An error of that magnitude in a material with significant embodied carbon would invalidate any comparative assertion between this building and a conventional alternative.

Contact: Site engineer / structural engineer on the Keur Songho project.

Item 2 — BLOCKED: BBSC composition (rows 15, 17, 32, 38)¶

Module location: quantities, all BoQ items using "BBSC on header course".

The problem: The workbook models BBSC bricks using only one material output: Rock = brick_count × 0.0002 m³/brick. No cement, no earth, no water, no binder is modeled for BBSC. The acronym "BBSC" is never expanded anywhere in the workbook.

Working hypothesis (not confirmed): BBSC = Blocs de Béton Stabilisé Comprimé (compressed stabilised concrete/earth block), a form of CSEB. If so, its true material composition includes at minimum: earth/aggregate, stabiliser (cement, lime, or pozzolan), and water. Treating a stabilised block as pure quarried rock understates its embodied cement/binder and may significantly misrepresent its GWP. This is the same structural failure as the slab rebar omission: a material is present in the building but absent from the LCA.

What we need from you: 1. What does "BBSC" stand for? 2. The material composition per brick: earth, cement (or other binder), aggregate, water (in m³ or kg per brick, or per m³ of finished block). 3. The source of the composition data (product datasheet, manufacturer, lab test, literature). 4. Whether the 0.0002 m³/brick rock figure in the workbook has any meaning (e.g. aggregate fraction), or is a placeholder. 5. Name and date of whoever can confirm the composition.

Why this is BLOCKED: We cannot compute the LCA impact of BBSC bricks without knowing what they are made of. Using the current rock-only model would make the LCA non-compliant with EN 15804+A2 §6.3.3 ("all significant materials shall be included").

Contact: Material supplier / procurement contact for Keur Songho.

Item 3 — EXPERT-JUDGEMENT: OMPU reference document¶

Module location: Throughout quantities — every conversion factor in the workbook.

The problem: Row 72's cell note explicitly states "Not in OMPU, brief proportions" for the Tyrolean plaster recipe. This implies that every other recipe in the workbook IS from "OMPU" — an acronym never expanded anywhere in the source material. Every factor currently marked PROVENANCE UNKNOWN in the spec's Factors table is presumably from OMPU.

What we need from you: 1. What is OMPU? (Full name, publisher, edition/date, where to obtain it.) 2. Is it a Senegalese BTP norms document, a company internal standard, or a university reference? Does it carry a licence that allows us to cite it? 3. Is the document available in digital form?

Why this matters: Under ISO 14071, every input parameter to a reported LCA result must have a traceable source. "PROVENANCE UNKNOWN" is acceptable during development but not in a final auditable result. Identifying OMPU would allow us to move the entire Factors table from UNKNOWN to SOURCED.

Status: EXPERT-JUDGEMENT (pending) — the tool will run and flag these as un-cited. The ISO 14071 reviewer will flag them.

Contact: Adama (referenced in the orphaned cell comment on row 43: "À vérifier avec Adama", dated 2024-02-08) or whoever assembled the original workbook.

Item 4 — EXPERT-JUDGEMENT: Anti-termite treatment reference area¶

Module location: quantities, BoQ item A.IX.2 — "Waterproofing and anti-termite treatment in the foundation."

The problem: The workbook computes U36 = 1 × 5 L/m² × 289.15 m², where 289.15 m² is the roof area (D35). Anti-termite treatment is applied to the foundation zone (soil under and around the slab), not to the roof. The correct reference area is the foundation contact area or the treated perimeter × treatment depth.

What we need from you: 1. The correct reference area for the anti-termite treatment on the Keur Songho project: - Foundation plan area (m²)? - Or: treated perimeter (m) × treatment depth (m)? 2. The product specification for the treatment (name, active ingredient, rate in L/m²). 3. Is the 5 L/m² rate confirmed from a product datasheet, or is it an approximation? 4. Name and date of whoever specified the treatment.

Status: EXPERT-JUDGEMENT (pending) — the tool currently uses roof area as the proxy. A divergence test (test_c5_anti_termite_area_attribution) is marked XFAIL until this is resolved.

Item 5 — EXPERT-JUDGEMENT: Steel density for fenestration¶

Module location: quantities, all window/door items (rows 60, 61, 63, 64, 65, 67).

The problem: Fenestration steel (window/door frame plates and profiles) is tracked by volume (m³) in the workbook, while structural rebar is tracked by mass (kg). There is no steel density conversion anywhere in the workbook, so the two steel streams cannot be summed or compared without assuming a density. The tool will need a density to convert fenestration steel volumes to mass for EPD impact factor application (EPD data is typically per kg of steel).

What we need from you: 1. The steel grade/type used for window and door frames (e.g. S235/S275 structural steel, or a lighter-gauge cold-formed section). 2. If you are comfortable approving a standard density: mild structural steel ≈ 7,850 kg/m³ (per EN 10025 / common reference). Can we use this? 3. Name and date of whoever can confirm.

Status: EXPERT-JUDGEMENT (pending) — the tool can use 7,850 kg/m³ if Marco approves it as the project-wide default, with named attribution in the audit trail.

Item 6 — EXPERT-JUDGEMENT: Sand density scope¶

Module location: quantities, row 72 (B.IV.4 — Tyrolean plaster).

The problem: Row 72's cell comment states "density of sand is considered → 1600 kg/m³". This is the only place in the workbook where a sand density is stated explicitly. The question is whether this value is specific to the Tyrolean plaster (a coarser plaster aggregate), or whether it should be used elsewhere in the workbook where sand mass ratios implicitly exist.

What we need from you: 1. Is 1600 kg/m³ the intended density for all sand in this project, or specific to row 72? 2. Source of the 1600 kg/m³ value (lab test, reference table, assumption)?

Status: EXPERT-JUDGEMENT (pending) — currently applied only in row 72, which is where it appears. No extension to other rows without your confirmation.

Item 7 — EXPERT-JUDGEMENT: Antirust paint — solvent-phase vs water-phase¶

Module location: lca.b4, lca.c3, lca.c4, lca.d (Phase 2 unbuilt modules). Material: PAINT_ANTIRUST (MaterialCategory).

The problem: The antirust paint GWP factor in 2_lca_vn.xlsx (105 kg CO₂/kg) is an uncited anomaly approximately 18–25× higher than any published EPD value. The provisional replacement from INIES is:

Solvent-phase alkyd (INIES FDES 2-152:2020, SIPEV): 4.87 kg CO₂/kg
Water-phase alkyd (INIES generic category): 3.23 kg CO₂/kg

The two values differ by ~34%. The choice between them depends on which product Chiwara actually specifies for its projects. This is a Chiwara material-classification decision.

What we need from you: 1. Is the antirust primer used on Keur Songho and Kalan Deme So projects solvent-phase (alkyde en phase solvant) or water-phase (alkyde en phase aqueuse)? 2. If you have a product name or supplier, please provide it (to allow direct EPD lookup). 3. Name and date of whoever specifies the antirust primer on Chiwara projects.

Note on BLOCKER 1: INIES FDES 2-152:2020 (SIPEV, solvent-phase) expired June 2025. Even if you confirm solvent-phase, Marco must locate a current FDES on INIES before the value is committed as a seed EPD. This item documents the type decision; the sourcing decision is separate (see antirust-paint-investigation.md §4 BLOCKER 1).

Status: EXPERT-JUDGEMENT (pending) — no antirust seed EPD can be committed until both this item and BLOCKER 1 are resolved.

Contact: Project architect or procurement contact for Keur Songho / Kalan Deme So.

Item 8 — BLOCKED: B4 replacement parameters (RSL, % Finishes, study period)¶

Module location: lca.b4 — Module B4 (Replacement), all materials.

Source: reference/excel/2_lca_vn.xlsx, sheet B4 (Replacement). Spec reference: docs/methodology/lca-stages-b4-c3-c4-d-spec.md §3.

The problem: The Excel workbook applies a single hardcoded RSL of 10 years to all materials, with a per-material "% Finishes" fraction determining what proportion of each material's installed quantity is replaced. Neither the RSL of 10 years nor the per-material finishes fractions have a normative source cited anywhere in the workbook. Decision #24 (PROJECT.md) requires per-material RSL values that are SOURCED (normative citation) or EXPERT-JUDGEMENT (named practitioner, date, basis).

What we need from you:

Study period: The workbook uses 50 years (sheet B1, cell C5). Decision #15 defaults to 30 years for the Sahel context. Which value is correct for the Keur Songho project? Please name and date the person confirming this.
RSL per material: For each of the following materials, what is the Reference Service Life (years), and what is the basis?

Material	Excel RSL used	Excel % Finishes	Question
Sand (render component)	10 yrs	50%	Is 50% of installed sand in renders/finishes? Source?
Earth (render component)	10 yrs	8%	Is 8% the render fraction? What RSL for earth render in BSh climate?
Water (maintenance)	10 yrs	3%	What maintenance activity uses 3% of BoQ water?
Cement (finishes)	10 yrs	40%	40% cement in finishes — which elements?
Bitumen (waterproofing)	10 yrs	100%	Waterproof membrane replaced every 10 years? Source?
Waste Oil	10 yrs	100%	Is waste oil a maintenance material? What is the activity?
Anti-termite	10 yrs	30%	30% of treatment requires re-application? At what interval?
Wall paint	10 yrs	100%	Repainting every 10 years — is this confirmed? Source?
Antirust paint	—	0%	No replacement modelled. Is this intentional (one-time application)?
Glass	—	0%	No replacement modelled. What is the RSL for the fanlights?
Floor tiles	—	0%	No replacement modelled. What is the RSL for school floor tiles in Sahel?

Crude oil: The workbook treats crude oil (60 kg) as a 100% maintenance material replaced 4 times. Crude oil in the BoQ appears to be a construction-phase fuel/lubricant (used in A5 or as form-work oil). Is it intentionally included in B4? If so, what is the maintenance activity?

Why BLOCKED: Without per-material RSL values from a named practitioner or normative source, lca.b4 cannot compute replacement cycles for any material. The module raises ValueError for any material where RSL is not attributed. An entire B4 module full of BLOCKED parameters fails ISO 14071 defensibility.

Contact: Project architect or structural engineer on Keur Songho; LCA methodology lead.

Item 9 — BLOCKED: Crude oil and waste oil scope in Module B4¶

Module location: lca.b4.

Source: reference/excel/2_lca_vn.xlsx, B4 (Replacement) sheet rows 16–17.

The problem: Crude oil (E16 = 1.0, 100% replacement) generates 32 221 kg CO₂ eq in B4 — 47.9% of the entire B4 total. This dominates the module. The workbook provides no rationale for including crude oil as a maintenance material. The LCI Final note for crude oil says "No EoL as it is burned (all emissions go in A1-3)" — yet B4 treats it as a recurring maintenance item. This is internally contradictory.

Waste oil (E17 = 1.0, 100% replacement) adds another 120 kg CO₂ eq.

What we need from you: 1. What maintenance activity uses crude oil on the Keur Songho project? (Examples: form-work oil replenishment, machinery lubrication, generator fuel?) 2. Is crude oil scoped to Module B4 (maintenance) or to Module A5 (construction)? 3. Similarly for waste oil — what maintenance activity generates waste oil, and at what frequency over the 50-year study period? 4. Name and date of whoever specified these B4 inclusions.

Verdict if crude oil confirmed as non-B4: CORRECTED (B4-F2) — crude oil B4 impact becomes 0.0. Total B4 reduces by ~32 221 kg CO₂ (−47.9%).

Contact: Site engineer / project manager for Keur Songho.

Item 10 — BLOCKED: C3/C4 disposal rates and GWP factors¶

Module location: lca.c3, lca.c4.

Source: reference/excel/2_lca_vn.xlsx, sheet LCI Final, columns G (disposal rate), J (C3 factor), K (C4 factor).

Spec reference: docs/methodology/lca-stages-b4-c3-c4-d-spec.md §4.3, §5.3.

The problem: All disposal rates (what fraction of each material goes to landfill/processing vs. recycling) and all C3/C4 GWP factors have no normative sources in the workbook. The only references are: "ÖKOBAUDAT" (no dataset UUID), "Pietro's Thesis" (not an EPD source), "ok" (not a citation), and "Poland company's EPD" (unnamed).

Specific concerns requiring your input:

Polyethylene film C3 factor (1.13 kg CO₂/kg): The unnamed "Poland company's EPD" drives 79.5% of the entire C3 total (5 679 kg CO₂). Can you identify this EPD (programme operator, EPD number, product name)? Without identification, this factor is BLOCKED.
Steel C4 factors (7.18 kg CO₂/kg for rebars/profiles, 1.81 for plates): These are anomalously high for landfill disposal of steel. For context, published landfill emission factors for inert metals are typically <0.01 kg CO₂/kg. Values of 7.18 would suggest incineration, not landfill. Can you identify the source?
Paint C4 factors (1.0 kg CO₂/kg for wall paint and antirust paint): Similarly high for landfill disposal. Source?
Earth disposal rate (8%) and C3 factor (0.003378 kg CO₂/kg): Sourced to "Pietro's Thesis." The Pedrazzi thesis is our methodology reference, not a primary EPD source. What is the underlying source Pietro used?
Anti-termite C3 factor (0.05 kg CO₂/kg): Same factor applied to wall paint and antirust paint — is this intentional? Source?

Contact: LCA methodology lead; Pietro Pedrazzi (for thesis primary sources).

Item 11 — BLOCKED: Module D recycling credits — net accounting basis¶

Module location: lca.d.

Source: reference/excel/2_lca_vn.xlsx, sheet LCI Final, column L (D factors).

Spec reference: docs/methodology/lca-stages-b4-c3-c4-d-spec.md §6.3.

The problem: For earth, rock, gravel, and sand, the Module D credit formula is L = −I (negative of the A1-A3 manufacturing factor). This is a 100% gross avoided-burden credit: it assumes the recycled material fully substitutes for virgin primary production with no secondary processing cost.

EN 15804+A2 §6.4.3.2 requires the net flows method: the credit must equal (primary production avoided) minus (secondary material production burden). For aggregate reuse as fill, the secondary burden may be near-zero (transport + compaction), making the gross approximation close to correct, but this must be documented.

For steel, the D credits (−1.48, −1.01, −1.48 kg CO₂/kg) are less than the A1-A3 factors (2.94, 1.65, 2.17 kg CO₂/kg), suggesting a pre-computed net credit. But no source is cited.

What we need from you: 1. For earth, rock, gravel: what processing does the recycled material undergo (crushing, screening, transport)? Is the secondary burden documented? 2. For steel: what is the source of the D credit values (−1.48, −1.01, −1.48 kg CO₂/kg)? Was a net accounting calculation performed? 3. Name and date of the LCA practitioner who assigned these values.

Contact: LCA methodology lead; Pietro Pedrazzi.

Item 12 — BLOCKED: Earth A1–A3 GWP factor — primary source required¶

Module location: lca.a1_a3, lca.b4 (earth render fraction).

Source: reference/excel/2_lca_vn.xlsx, LCI Final!I6 = =4.675/1000 = 0.004675 kg CO₂/kg. Note in workbook: "Pietro's Thesis". Spec reference: docs/methodology/material-impact-database-spec.md §3.4.

The problem: The earth GWP factor of 0.004675 kg CO₂/kg originates from the Pedrazzi thesis (MSc dissertation, Kalan Deme So, Mali). The thesis is a methodology reference document — it summarises existing literature but is not itself a primary EPD source. The underlying sources the thesis cites must be identified and linked.

Under PROJECT.md decision #26, Earth is a locally-produced EARTH material for which no ÖKOBAUDAT or ICE substitute is permitted. The primary source must be either: - Chiwara field measurements of earth extraction energy and transport for Keur Songho / Kalan Deme So; - CRAterre published data for the specific earth type used (laterite, alluvial, etc.); - RILEM TC 274-TCE data for comparable Sub-Saharan Africa earth; - Ben-Alon et al. (2020) peer-reviewed values for adobe/CEB earth in the relevant climate zone.

lca.a1_a3 raises ValueError for EARTH / RAW_EARTH until this is resolved.

What we need from you: 1. Where does the 0.004675 kg CO₂/kg value come from? Can you identify the primary source in the Pedrazzi thesis (specific reference, table, page)? 2. What type of earth is used in Keur Songho (laterite, alluvial clay, other)? Where is it extracted relative to the site (distance, extraction method)? 3. Is there a CRAterre, RILEM, or published EPD that applies to this specific earth type? 4. Name and date of whoever can confirm the earth specification and sourcing.

Contact: LCA methodology lead; Pietro Pedrazzi; CRAterre collaborators.

Item 13 — BLOCKED: Crude oil A1–A3 = 134 kg CO₂/kg — anomaly investigation¶

Module location: lca.a1_a3.

Source: reference/excel/2_lca_vn.xlsx, LCI Final!I14 = literal 134. Note in workbook: "No Eol as it is burned (all emissions go in A1-3)". Spec reference: docs/methodology/material-impact-database-spec.md §3.12.

The problem: 134 kg CO₂/kg for crude oil is approximately 44× higher than the combustion emission factor for crude oil (IPCC EFDB Tier 1: ~3.0 kg CO₂/kg including both extraction and combustion). The workbook note says "burned in A1-3" (construction-site fuel), but even including full combustion, 134 kg CO₂/kg is not derivable from any published LCA methodology.

This appears to be the same class of anomaly as the antirust paint (see antirust-paint-investigation.md): a factor that is orders of magnitude higher than any published value with no traceable source. The most likely root cause: 134 MJ/kg is the approximate net heating value (calorific value) of crude oil — this energy value was likely imported into the GWP factor cell by mistake.

The crude oil A1-A3 factor generates 8 055 kg CO₂ — 8.7% of the A1-A3 total. If corrected to a typical combustion factor (~3.0 kg CO₂/kg), the contribution would be ~180 kg CO₂ — a reduction of ~7 875 kg CO₂ (−8.5% of total A1-A3).

lca.a1_a3 raises ValueError for PETROLEUM_DISTILLATE until the factor is investigated and a primary source identified.

What we need from you: 1. What is the primary source for the crude oil GWP factor = 134 kg CO₂/kg? Was it taken from a published reference, or computed from a formula? 2. Was "134" perhaps entered as an energy value (134 MJ/kg = crude oil heating value) rather than a GWP factor? 3. Is crude oil used on the Keur Songho project as a construction-phase fuel (A5 scope), or is it a material in the building (in which case its GWP must include combustion products)? 4. If the intent is to capture combustion emissions, the correct factor is approximately 3.0 kg CO₂/kg (IPCC EFDB crude oil, Tier 1). Is this acceptable as a provisional correction? 5. Name and date of whoever assigned the 134 value.

Contact: LCA methodology lead; workbook author.

Item 14 — BLOCKED: Floor tiles A1–A3 = 5.09 kg CO₂/kg — suspected unit error¶

Module location: lca.a1_a3, lca.b4 (B4-F6 CORRECTED: tiles should have RSL-based replacement).

Source: reference/excel/2_lca_vn.xlsx, LCI Final!I20 = literal 5.09. No source note in workbook. Spec reference: docs/methodology/material-impact-database-spec.md §3.18.

The problem: 5.09 kg CO₂/kg for fired ceramic floor tiles is 26.5× higher than the ICE published value (0.192 kg CO₂/kg) and well outside the published range for ceramic tiles (0.10–0.80 kg CO₂/kg for standard tiles; up to 1.5 kg CO₂/kg for high-fired porcelain).

This appears to be a unit error: 5.09 may be in kg CO₂/m² (a common EPD declared unit for floor finishes), not per kg. If corrected: - At 10 mm tile thickness and 2400 kg/m³: 5.09 ÷ (0.010 × 2400) = 0.212 kg CO₂/kg — consistent with ICE (0.192) and within the published range. - The glass factor in the same workbook shows an explicit declared-unit → per-kg conversion (formula =9.71*(1/0.006)*(1/E16)). Floor tiles lack a similar conversion, suggesting the EPD value was applied directly without unit adjustment.

Note also that the weight formula LCI Final!F20 = =(C20*0.005)*E20 uses a 5mm thickness factor (not 10mm). If the EPD declared a 10mm standard tile, this adds a second layer of confusion.

lca.a1_a3 raises ValueError for CERAMIC_TILE until the EPD and declared unit are confirmed.

What we need from you: 1. What is the source EPD (programme operator, EPD number, product name) for the floor tile GWP factor? 2. What is the declared unit of the EPD: per kg, per m², per m² at stated thickness? 3. What tile thickness is assumed in the project? (The weight formula uses 5mm — is this correct for the tiles specified at Keur Songho?) 4. Name and date of whoever selected the tile EPD.

Contact: LCA methodology lead; procurement contact for Keur Songho.

Item 15 — BLOCKED: Raw-earth primary data package (density + A1–A3 GWP)¶

Module location: lca.a1_a3 (EARTH / RAW_EARTH), lca.a4, lca.b4 (earth render fraction), and the Keur Songho mass path in backend/app/domains/lca/a1_a3.py.

Register rows: AR-47 (this item), with AR-03 (earth factor) and AR-46 (mass path).

Why this is one request and not two. An impact is E = weight × factor. For earth we currently have neither half from a citable source:

The factor (A1–A3 GWP) traces only to "Pietro's Thesis" — see Item 12. That is a methodology reference, not a primary source.
The weight is back-solved. The workbook's earth mass, 1 494 888.525 kg, divided by the workbook's earth volume, 830.5 m³, gives exactly 1 800 kg/m³ — a round number that appears nowhere as a stated, sourced density. It inherits the OMPU provenance gap (Item 3). And the 830.5 m³ is itself pre-correction: the corrected vault geometry gives 751.9 m³ (−9.5%), so the mass path has to be rebuilt from corrected volumes × a real density regardless.

Answering only one half does not produce a citable number. That is why both are asked here together.

What we need from you:

Bulk density of the Keur Songho raw earth, as placed (kg/m³) — i.e. the compacted in-wall density, not loose or quarry-face density. Please state which it is.
Basis: lab test, site measurement, or a CRAterre / published reference. Please give the method and the date.
If a single project-wide value is not appropriate (rammed earth vs adobe vs vault brick differ), give the value per application.
A1–A3 GWP for local earth extraction and preparation (kg CO₂-eq/kg or per m³ — state which), with a citable source. Acceptable sources are:
a Chiwara primary measurement (fuel or machine-hours for extraction, haulage to the preparation area, mixing/screening energy — raw measurements are fine, we can compute the factor from them);
CRAterre published data for the specific earth type;
RILEM TC 274-TCE, or Ben-Alon et al. (2020), for comparable Sub-Saharan earth.
The same two figures for the BBSC binder and the adobe / BBSC production process, if primary data exists: binder type and dose, block density, and the energy or fuel used in block production. This also feeds Item 2 (BBSC composition).
Name and date of whoever can confirm each figure.

No substitute is permitted. PROJECT.md decision #26 blocks ÖKOBAUDAT — and every other European dataset — for locally-produced earth, and lca.a1_a3 enforces this as a hard block on the LCI Final!I6 factor. We cannot fall back to a European proxy and flag it; the module raises. This data can therefore come only from Chiwara or CRAterre.

If it does not exist, that is the answer we need. If no measurement was ever taken and no CRAterre figure covers this earth type, please say so plainly. A documented "no primary data exists for the Keur Songho earth" is itself a finding: it is recorded as a BLOCKED parameter in the audit trail and, if we go on to commission a measurement, it is the justification for doing so. It is not a failure to answer.

Contact: Site engineer / project manager for Keur Songho; CRAterre collaborators; Pietro Pedrazzi for the thesis chain.

Item 16 — Confirmation requested: mixing water excluded by EN 15804 cut-off¶

Module location: lca.a1_a3 (WATER).

Register row: AR-48.

Source: reference/excel/2_lca_vn.xlsx, LCI Final!I7 = literal 0, note "NA". Spec reference: docs/methodology/material-impact-database-spec.md §3.5.

The situation: Keur Songho uses 299 583.55 kg of water (mixing, curing, block production). The workbook assigns it an A1–A3 GWP of zero, and our spec accepts that as standard LCA convention — potable-water processing emissions are negligible at these volumes, and any transport impact belongs to A4. We are not disputing the number.

What is missing is the status of the exclusion. Right now it is an undeclared convention. Under EN 15804+A2 §6.3.3 it should be a declared cut-off, with a party who confirms it is appropriate for the comparison being made. An ISO 14071 reviewer will ask who decided, and when.

What we need from you: confirmation, in one line, that a formal EN 15804 cut-off exclusion of mixing water is acceptable for the earth-vs-conventional comparison — that is, that no comparative assertion between the earth building and a conventional alternative turns on the water term. Plus your name and the date.

Why it is asked here rather than in passing: it is a low-stakes confirmation, and it would be easy to settle in a phone call. But an answer given off-channel leaves no audit trail, and this document is the audit trail. Everything that changes the status of a parameter goes through it.

Status: EXPERT-JUDGEMENT (pending) — water computes at exactly 0 today and will continue to. Your confirmation changes how it is described in the audit trail, not what it computes.

Contact: LCA methodology lead; project architect for Keur Songho.

Open question: OMPU water:earth ratio for vaults (K29, K30)¶

The vault water formula uses a derived ratio 0.2585 m³ water / m³ earth, calculated from brick dimensions (0.0043 ÷ (0.42 × 0.22 × 0.18)). The brick dimensions (0.42 m × 0.22 m × 0.18 m) are stated in-sheet but have no external citation. Are these the standard Nubian vault brick dimensions from a CRAterre or Voûte Nubienne Association specification, or project-specific dimensions?

If these come from a published standard, that would allow the water:earth ratio to be classified as SOURCED rather than EXPERT-JUDGEMENT. Please advise.

How to respond¶

Please fill in answers inline under each item above, add your name and date, and return this document to the LCA team. For BLOCKED items, a response is required before the tool can compute material quantities for the Keur Songho reference project.

For questions, contact: marco.ferrario@gmail.com