Consequential Life Cycle Assessment of pisco production in...
Transcript of Consequential Life Cycle Assessment of pisco production in...
Life Cycle Management Conference – LCM 2017 Gustavo Larrea-Gallegos
Ian Vázquez-Rowe
Ramzy Kahhat
Hugo Wiener
Consequential Life Cycle Assessment
of pisco production in the Ica Valley,
Peru
Luxembourg
September 6th 2017
Presentation layout
Pisco production as a case study
Methods and model development
Results
1
2
3
Conclusions and discussion4
Pisco is an alcoholic beverage produced with grapes harvested from
southern coastal Peru
City of Ica
City of Pisco
Viticulture was identified as a very intensive stage in terms of climate
change in most wineries
Viticulture
Vinification
Distribution
ALCA boundaries
CLCA boundaries
Crop dynamics
Technological dynamics
Factor of scale
Market
dynamics2,99
1,97
kg o
f C
O2eq
per
lite
ro
f p
isco
Contribution of stages in GHG emissions per liter of pisco
Viticulture Vinification
Vázquez-Rowe, I., Cáceres, A. L., Torres-García, J. R., Quispe, I., & Kahhat, R. (2017). Life Cycle Assessment
of the production of pisco in Peru. Journal of Cleaner Production, 142, 4369-4383.
1,60 1,801,50
2,402,90
3,90
4,90
6,106,50 6,60
6,30 6,30
7,10 7,10
8,60
0,00
1,00
2,00
3,00
4,00
5,00
6,00
7,00
8,00
9,00
10,00
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
MIL
LIO
N O
F LI
TER
S
YEAR
PISCO PRODUCTION 2000-2014
It is reasonable to expect an increase in the demand of pisco for the
following years with BAU conditions
BAU: Business as usual
14.3
16.8
2025 2030
It is reasonable to expect an increase in the demand of pisco for the
following years with BAU conditions
BAU: Business as usual
Increase in demand of
pisco
Increase in demand of
grapes
Replacement Expansion Imports
Profit of grape increases during time and is always
higher than for certain crops
Intensification
Fertilization and water use may
alter grape quality
The objective is to understand the environmental consequences of
the upcoming increase in pisco demand in terms of GWP
Focusing only on
agriculture stage
It is not replaceable due to
its denomination of origin
H 2: A given farmer will always choose
the most profitable crop
H 3: There are certain crops for which
profit remains constant or decreases
through time
H 1: Profit of grape increases during time
and is always higher than for certain
crops
Replacement and expansion scenarios can be explained by proposing
plausible economic conditions
𝑢 = 𝑓(𝑡)
𝑢1 = 𝑔(𝑡)
𝑢2 = ℎ(𝑡)
𝑢3 = 𝑘(𝑡)
pro
fit
time (demand)
Profit behavior given a demand Y
𝑢 − 𝑢1
𝑢 − 𝑢2
𝑢 − 𝑢3
𝑚𝑎𝑥𝑍 = 𝑢 − 𝑢1 𝑋1+ 𝑢 − 𝑢2 𝑋2+ 𝑢 − 𝑢3 𝑋3
𝑚𝑎𝑥𝑍 = 𝑢 𝑋1 + 𝑋2+ 𝑋3 − (𝑢1𝑋1+ 𝑢2𝑋2+ 𝑢3𝑋3)
𝑚𝑎𝑥𝑍 = 𝑢𝑌 − (𝑢1𝑋1+ 𝑢2𝑋2 + 𝑢3𝑋3)
𝑚𝑎𝑥𝑍 = −(𝑢1𝑋1+ 𝑢2𝑋2+ 𝑢3𝑋3)
𝑚𝑖𝑛𝑍 = 𝑢1𝑋1+ 𝑢2𝑋2+ 𝑢3𝑋3
Producer
perspective
Seasonal crops
Domestic sales
Low profit products
Financial
perspective
Total arable land > 3 ha
Arable land for grapes > 3 ha
Access to water sources
Selection criteria for marginal producers
National Agriculture Census 2012 and National
Production Database
A Stochastic Technology Choice Model was selected to analyze the
interaction among crops in the Ica and Pisco valleys
“n” sub-systems
columns
“n” sub-systemoutput rows
“m-n” replaceable crops
𝐴𝑛 𝑥 𝑛 [ ] 𝑛 𝑥 (𝑚−𝑛)
“k” arearequirement rows
𝐴∗𝑛 𝑥 𝑚
𝐹𝑘 𝑥 𝑚
“k” crop profit
elements𝑃𝑘𝑥1
“k” available land
elements𝐶𝑘𝑥1
“n” demand
elements𝑌𝑛𝑥1
min 𝑍 = 𝑃′𝐹𝑠
s. t. 𝐴𝑠 = 𝑌
𝑠𝑗 ≥ 0𝐹𝑠 ≤ 𝐶
Linear programming model:
Katelhon, A., Bardow, A., & Suh, S. (2016). Stochastic Technology Choice Model for Consequential Life Cycle
Assessment. Environmental Science & Technology, 50(23), 12575-12583.
Reinout, H., & Sangwon, S. (2002). The computational structure of life cycle assessment. The
Netherlands: Kluwer Academic Publishers.
There is an increase in the net GHG emissions per liter of pisco in
both scenarios due to crop replacement
0
1
2
3
4
5
6
7
8
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
kg C
O2eq
per
lite
r
rep
lace
d c
rop
2030 demand y = 8240000
6.21
0
1
2
3
4
5
6
7
8
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
kg C
O2eq
per
lite
r
rep
lace
dcr
op
2025 demand y = 5660000
Valley of Ica scenario Valleys of Pisco and Ica scenario
6.78
A profit variability simulation shows that emissions per FU converge
to a value and net emissions do not differ
GHG emissions in 2025
Ica – with or without annual variability:
38.4 Mg CO2eq
It is possible to satisfy the projected pisco demand in the Region of
Ica, if land and water are used efficiently
It is plausible for farmers to change to more profitable and less water demanding
crops (grapes)
The Valley of Pisco can potentially fulfill current and future pisco grapes
demand and minimize crop replacement
Fallow land is so abundant that a possible agricultural expansion is unlikely to
occur
Acknowledgement and contact information
• Dirección General de Investigación (DGI), from the PUCP, for funding this project
• Juan Pablo Bentín, from Valle y Pampa Agroindustrial Company, for the consulting
Gustavo Larrea-Gallegos :
Ian Vázquez-Rowe :
Ramzy Kahhat :
Hugo Wiener :