HWD Difference in Steel Hot Rolled Coils vs steel low alloyed from ecoenvent.

Question

Hello everyone,

I am trying to model an LCA for a product that contains a large amount of steel. However, when I use the following ecoinvent flow and provider:

Flow: steel, low-alloyed, hot rolled
Provider: steel production, low alloyed, hot rolled | steel, low alloyed, hot rolled | EN15804GD, U-RoW

I get a significantly higher HWD (Hazardous Waste Disposal) and NHWD (Non-Hazardous Waste Disposal) compared to the EPD of AM/NS – 1 tonne Steel Hot Rolled Coils.

The difference is quite large. I am trying to understand what might be causing this difference.

Am I using the wrong dataset/provider or system model? Is there a more appropriate dataset for comparing HWD and NHWD with the AM/NS EPD?

Per kg steel	ecoinvent result	AM/NS EPD (A1-A3) (EPD-IES-0012007)
HWD	2.58 kg	7.7E-10 kg
NHWD	9.5 kg	0.027 kg

Answer 1

1E-10 kg of hazardous waste for 1 ton of steel is really really low I think. A simple google answer e.g. says this: 

I looked for the EPD and it even does not have any biogenic GWP (so they assume no cardboard or wood for packaging is used - downloaded from here https://api.prod.environdec.com/api/v1/EPDLibrary/Files/EPDs/d344b5f9-5084-4049-c36b-08dbfcbd856b/Documents).

Waste needs to be specifically modeled (in the database and / or foreground system) otherwise it is "evened out" in the calculation (not being an elementary flow), and I assume people are often happy if a low result is obtained without questioning whether it makes sense. And here, I would say, it doesn't, even though it is a verified EPD.

Hope this helps!

Andreas

edit: here results from our latest EN15804 database - not 2.58 kg per kg but 0.053 kg per kg (for one example steel):

I admit there are cases with higher values (0.61 kg for steel production w. silicon alloy) but this is then coming from the ecoinvent database and we need to live with that I think, for generic background data..

Comments

I think this was an issue in our earlier versions, I post a picture above (editing my post) with results from our EN15804 database, latest version (with ecoinvent 3.12) - here the  amounts seem in order.