consumption up to the exit-gate of the grow-out farm.
However, the nutrient discharge of processing based on
Pham et al. (2010) is estimated at 16.5 kg N and 2 kg P per
ton fish processed (the yield of fillet is close to 40%). If
processing had been included in the LCA reported in this
paper, N output would have increased by about 60% and P
output would have increased by about 25%. The stakeholders
with an interest in aquaculture EIAs and LCAs
should be made aware of this and should consider including
the processing in future EIAs and LCAs.
Emissions of CO2, N2, N2O, NH3 and CH4 from the
pond surface to the air were ignored because of the lack of
sound data and because of the difficulties in quantifying
these emissions (they depend on the pH and dissolved
oxygen content of the pond and both fluctuate greatly).
Emissions from rice paddies have been reported, but
conditions at the deeper pond bottom are more anoxic
(e.g. methane production might be higher if not absorbed
in the water column). Research is needed to quantify the
airborne emissions from Pangasius ponds.
Photochemical oxidant formation and ozone depletion
were also ignored for the same reasons. Myrvang (2006)
and Mungkung (2005), however, showed that some cradleto-
gate processes contribute to these categories (e.g.
through energy use, transport, chemicals and medicines).
We may assume these impacts will reduce concomitantly
with the mitigation of impacts on GW and various toxicity
impacts.