Transcript Dr. Theodoros Zachariades, Ass. Professor

Assessment of Climate Change
Impacts on Electricity Consumption
and Residential Water Use in Cyprus
Theodoros Zachariadis
Dept. of Environmental Science & Technology
Cyprus University of Technology
tel. 25 002304, e-mail: [email protected]
November 2011
Analysis of Climate Effects on Energy Use
• Econometric time series analysis of energy use
in Cyprus by sector and fuel, 1960-2007
 Energy consumption = f (income/economic activity,
energy prices, time trends, weather)
• Climate effects captured by the variables of heating &
cooling degree days (they express intensity + duration
of cool & hot days respectively)
• Methodology: stationarity (unit root) tests of all
variables, application of cointegration techniques,
Vector Error Correction Models and Autoregressive
Distributed Lag Models
• Effect of climate statistically significant only for
electricity consumption in households & tertiary sector
Forecast of electricity consumption
up to 2030 – without climate change
Final electricity consumption (GWh)
16000
14000
Electricity use triples
by 2030
12000
10000
8000
6000
95% confidence intervals
Increased share of
domestic & tertiary
sector – 86% in total
4000
2000
0
1960
1970
1980
1990
2000
2010
2020
2030
Forecast of electricity consumption
up to 2030 – with climate change
• Assumption: uniform temperature increase by 1C in
2030, during the whole year
• Electricity use in 2030 higher by 2.9% (compared to ‘no
climate change’ scenario)
• Direct cost: 15 ΜEuros in 2020, 45 MEuros in 2030
• Present value of total cost in period 2008-2030:
> 200 MEuros (at constant prices of year 2007)
• Average cost per household: ~30 Euros/year in 2020,
~80 Euros/year in 2030 (at constant prices of year 2007)
• Further econometric analysis + forecast of peak
electricity load in summer with climate change:
additional 65–75 MW in 2020, 85–95 MW in 2030
 Increased requirements for extra reserve capacity
Work in progress
• Updated electricity consumption forecasts for
2030
• New forecasts for year 2050
– with new macroeconomic & price assumptions
– using recent climate change forecasts
(Hadjinicolaou et al. Regional Environmental Change, pp. 1-17, 10/2010)
Methodology to assess costs of water
shortages in non-agricultural sectors
Willingness to pay for water p (€/c.m.)
Water
demand curve
Welfare losses of
consumers due to reduced
availability of water
Price
q'
q0
Water quantity q
Estimating Residential Water Demand in Cyprus
• Data from the three Water Boards of Cyprus serving
the main cities (Nicosia, Limassol, Larnaca):
– Billed water consumption per consumer type
(residential, commercial, industrial)
– No. of consumers by type
– Water tariffs (fixed prices & prices per consumption block)
– Fraction of consumers in each consumption block
– Revenues and expenditures (from Board financial accounts)
– Period: 1980-2009 (annual data), 2000-2009 (data available
per billing period – 2/3/4 months)
• Other data:
– Monthly temperature and rainfall (from Met. Service)
– Quarterly GDP & population (from Statistical Service)
– Household income by district of Cyprus (Family Expenditure
Surveys conducted by Statistical Service)
Residential Water Demand Model
qit = f (incit , pit , pfixit , tempit , rainit , dummyi)
i: district (i = 1 to 3);
t: 4-month period from 2000/1 to 2009/3
q: water consumption per household (c.m.)
inc: household income (€)
p: water price (€/c.m.)
pfix: fixed part of water tariff (€)
temp, rain: temperature & rainfall level (C, mm)
dummy: for the period of interruptions in water supply
in each city (April 2008 – December 2009)
• Linear function, variables in logarithms, each
variable’s coefficient expresses an elasticity
• Two models:
a) p = average price, b) p = marginal price
Econometric Estimation
• Typical problem with block pricing: endogeneity of
prices – each consumer faces a water price that
depends on the quantity consumed.
Usual estimation method (OLS) will be biased
 Two-Stage Least Squares (2SLS) Estimation is
appropriate
• Requires identifying instruments that correlate with price
but not with dependent variable (water consumption)
• Three instruments were used:
• a) Consumer Price Index of previous year
• b) Water Board expenditures in previous year per c.m. of
water sold
• c) Same expenditures in current year
Estimation Results
Average price model
Coefficients of:
Income
0.529
Avg. price
-0.248
Fixed tariff -0.441
Temperature 0.241
Rainfall
0.047
Marginal price model
Coefficients of:
Income
0.753
Marg. price -0.449
Fixed tariff -0.490
Temperature 0.292
Rainfall
0.061
Dummies:
Nicosia
Larnaca
Limassol
Dummies:
Nicosia
Larnaca
Limassol
-0.034
0.065
-0.193
-0.025
0.088
-0.253
N = 73
Coefficients in bold are statistically significant at 1% level
Cyprus:
Costs of Water Shortages up to 2030
Scenario 1: Constant per capita water use
Water consumption
Cost
(mio c.m.)
(mio Euros'2009)
Year
2010
54.8
0.21
2015
57.0
0.75
2020
58.5
1.27
2025
59.5
1.73
2030
60.1
2.01
Total economic loss, 2010-30
25.57
Present value of economic loss, 2010-30
15.20
Scenario 2: Per capita water use grows 1% p.a.
Water consumption
Cost
(mio c.m.)
(mio Euros'2009)
54.8
0.21
60.0
1.95
64.6
5.15
69.1
9.86
73.3
15.84
130.69
71.96
Scenario 3: Per capita water use grows 2% p.a.
Water consumption
Cost
(mio c.m.)
(mio Euros'2009)
54.8
0.21
63.0
3.84
71.3
12.81
80.1
29.12
89.3
54.80
381.97
204.21
Additional scarcity cost due to climate change (mio Euros'2009)
Scenario 2: Per capita Scenario 3: Per capita
Difference in water
Scenario 1: Constant
Year availability due to
water use grows 1%
water use grows 2%
per capita water use
climate change
p.a.
p.a.
2010
0.0%
0.00
0.00
0.00
2015
-0.9%
0.17
0.28
0.41
2020
-1.9%
0.46
1.00
1.74
2025
-2.8%
0.85
2.27
4.58
2030
-3.7%
1.28
4.17
9.72
Total additional economic loss, 2010-30
11.11
29.42
60.19
Present value of economic loss, 2010-30
6.12
15.69
31.49
An adaptation measure: ‘Efficient’ household
water prices to account for scarcity
Year
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
λ (€cents'2009)
q i , without climate change (mio c.m.)
Constraint: Q  1095 mio c.m.
Scenario 1
Scenario 2
Scenario 3
50.8
46.9
43.1
51.2
47.7
44.2
51.5
48.3
45.1
51.7
48.9
46.0
51.9
49.5
47.0
52.1
50.1
47.9
52.3
50.7
48.9
52.4
51.2
49.7
52.5
51.6
50.6
52.5
52.1
51.4
52.6
52.5
52.3
52.6
53.0
53.2
52.6
53.3
54.0
52.6
53.7
54.8
52.5
54.1
55.7
52.5
54.5
56.5
52.4
54.8
57.4
52.3
55.1
58.1
52.2
55.3
58.9
52.0
55.6
59.6
51.8
55.9
60.4
27.0
62.8
113.9
Scenario 1
50.1
50.5
50.7
50.9
51.1
51.3
51.5
51.5
51.5
51.6
51.6
51.6
51.6
51.5
51.5
51.4
51.3
51.1
51.0
50.8
50.6
32.7
q i , with climate change (mio c.m.)
Constraint: Q  1075 mio c.m.
Scenario 2
46.2
47.0
47.6
48.1
48.7
49.3
49.8
50.3
50.7
51.1
51.5
52.0
52.3
52.7
53.0
53.4
53.7
53.9
54.2
54.4
54.6
70.8
Scenario 3
42.4
43.5
44.4
45.3
46.2
47.1
48.0
48.8
49.7
50.5
51.4
52.2
53.0
53.8
54.6
55.4
56.2
56.9
57.7
58.4
59.1
125.3
Climate change increases water shortages modestly, requires
8-13 €cents/c.m. higher water prices to induce conservation in
order to address this additional scarcity
An adaptation measure: Effects of
‘efficient’ household water pricing
(million cubic metres per year)
75
70
65
Unconstrained water consumption
Water conservation to be
achieved through scarcity pricing
at 63 €cents'2009 per c.m.
60
55
Average annual water availability
50
45
Constrained water consumption
40
2010
2015
2020
Year
2025
2030
An adaptation measure: What if we had
‘efficient’ water pricing already in 2000?
(million cubic metres per year)
75
Water conservation that would have
been achieved through scarcity
pricing at 25 €cents'2009 per c.m.
70
65
60
Unconstrained water consumption
55
50
Average annual water availability
45
Constrained water consumption
40
35
2000
2005
2010
2015
Year
2020
2025
2030