Monte Ca rlo Ana lysis of Monte Ca rlo Ana lysis of Unc e rta - - PowerPoint PPT Presentation
Monte Ca rlo Ana lysis of Monte Ca rlo Ana lysis of Unc e rta intie s in the Ne the rla nds Unc e rta intie s in the Ne the rla nds NIR for 1990- - 2004 2004 NIR for 1990 Je roe n va n de r Sluijs (UU) Andre a Ra mire z Ra mire z (UU)
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Je roe n va n de r Sluijs (UU) Andre a Ra mire z Ra mire z (UU) Corry de Ke itze r (UU) Jos O livie r (MNP) L a ure ns Bra nde s (MNP)
Se pte mbe r 27-28, 2007
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Asse ss whe the r a Mo nte Ca rlo a na lysis o f the unc e rta intie s in the Dutc h NI R wo uld re sult in diffe re nt le ve ls o f unc e rta intie s c o mpa re d to tho se pro vide d b y the T ie r 1 a na lysis. Asse ss whic h pa ra me te rs c o ntrib ute the mo st to the to ta l unc e rta inty in the e missio ns, in o rde r to ide ntify a re a s o f hig h prio rity fo r the furthe r impro ve me nt o f the
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Ana lysis pe rfo rme d fo r the K yo to b a se ye a r (1990/ 1995) a nd fo r 2004 Ba sic da ta fo r the c a lc ula tio ns o b ta ine d fro m NL NI R Sa me le ve l o f a g g re g a tio n c ho se n a s in NL T ie r 1 a na lysis
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
All kno wn no n-Ga ussia n distrib utio n we re ta ke n into a c c o unt No rma l distrib utio n a ssume d whe re sta nda rd de via tio n e q ua l o r b e lo w 30% L
de via tio n > 30% I n so me c a se s, no rma l distrib utio ns ha ve b e e n trunc a te d (0,+inf)
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
So ftwa re pa c ka g e use d: @ Risk PDF s fo r e missio n fa c to rs a nd a c tivity da ta we re b a se d o n unc e rta inty ra ng e s use d in T I E R 1 a nd e xpe rt judg me nt fro m MNP. All kno wn c o rre la tio ns we re ta ke n into a c c o unt
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
ACT I VI T Y DAT A: F OR T HE SAME YE AR AND DI F F E RE NT E MI SSI ON F ACT ORS SAME E MI SSI ON F ACT OR DI F F E RE NT SE CT ORS AND/ OR DI F F E RE NT YE ARS
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
F OR E MI SSI ON F ACT ORS F ROM SAME CAT E GORY AMONG YE ARS:
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
F
c a te g o rie s a mo ng ye a rs F
(e .g ., g a s c o nsumptio n in c he mic a l industry in c a te g o ry 1A a nd c he mic a l fe e dsto c k in c a te g o ry 2) Ac tivity da ta a mo ng ye a rs c o nside re d inde pe nde nt (CORR=0), E xc e ptio n: E missio ns fro m L a ndfill wa ste
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Following the IPCC Tier 2 method, uncertainties in the trend emissions were calculated in absolute and in relative terms. A pedigree assessment was carried out for the most sensitive emissions factors and activity data to systematically assess strengths and weakness in the knowledge base (part of the NUSAP system).
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Source: Risbey et al., 2001
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Tier 1: 1990: 213 Tg CO2eq 2004: 217 Tg CO2eq std: 4.5% (6%)
T
T
Without L UCF Without L UCF
1990
Mean = 214434.4 X <=205692.83 2.5% X <=227374.2 97.5% 1 2 3 4 5 6 7 8 9 190 218 245 273 300
Values in Thousands Values in 10^ -5 2004
Mean = 217211.1 X <=226654.06 97.5% X <=210429.5 2.5% 0.0 0.2 0.4 0.6 0.8 1.0 1.2 200 220 240 260 280
Values in Thousands Values in 10^ -4
Monte Carlo analysis-: 1990: 214 Tg CO2eq 2004: 217 Tg CO2eq std: 5.3 % (1990); 3.9% (2004) 1990 2004
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Tier 1: Relative change: 1.6% std: 3.3%
Monte Carlo analysis: Relative change: 1.3% std: 4.5%
Without LUCF
Mean = 2776.677 X <=-7650.37 2.5% X <=11529.15 97.5%
1 2 3 4 5 6 7 8 9
20 40
Values in Thousands Values in 10^ -5
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Year Minimum Maximum Mean Mode Std Deviation Variance Skewness Kurtosis 1990 154120.8 164436.3 158974.5 158297.1 1199.3 1438225.0 0.043 2.986 2004 173637.8 185306.1 179516.0 177994.8 1318.0 1737040.0 0.034 3.011
1990
Mean = 158974.5 X <=161351.22 97.5% X <=156659.83 2.5%
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 154 158 162 166
Values in Thousands Values in 10^ -4 2004
Mean = 179516
X <=182112.19 97.5% X <=176954.98 2.5% 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 172 176 179 183 186
Values in Thousands Values in 10^ -4
Tier 1: 1990: 159 Tg CO2eq 2004: 179 Tg CO2eq std: 2.5% (5%)
2 emissions
Monte Carlo analysis 1990: 159 Tg CO2eq 2004: 180 Tg CO2eq std: 1.5 % (1990 & 2004) 1990 2004
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Tier 1: Relative change: 9.6% std: 2.1%
2 emissions
Monte Carlo analysis: Relative change: 9.6% std: 1.6%
M ean = 20541.49 X < =23579.72 97.5% X < = 17501.92 2.5%
0.0 0.5 1.0 1.5 2.0 2.5 3.0 12 16 20 24 28
V alues in T housands Values in 10^ -4
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Year M inimum M aximum M ean M
Std Deviation Variance Skewness Kurtosis 1990 16889.97 38036.09 25463.84 25716.46 2381.692 5672456 0.4433465 3.979612 2004 12807.72 29766.56 17444.48 17203.42 1315.159 1729642 0.3799421 3.685013
1990
Mean = 25463.84 X <=30988.2 97.5% X <=21042.94 2.5%
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 15 21 28 34 40
Values in Thousands Values in 10^ -4 2004
Mean = 17444.48 X <=20236.8 97.5% X <=15068.76 2.5%
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 12 18 24 30
Values in Thousands Values in 10^ -4
Tier 1: 1990: 25.4 Tg CO2eq 2004: 17.5 Tg CO2eq std: 18% (25%)
4 emissions
Monte Carlo analysis -: 1990: 25.5 Tg CO2eq 2004: 17.4 Tg CO2eq std: 18.7 % (1990); 15.1% (2004) 1990 2004
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Tier 1: Relative change: -3.7% std: 1.4%
4 emissions
Monte Carlo analysis Relative change: -3.7% std: 2.2%
Mean = -8019.355 X <=-3370.86 97.5% X <=-13540.66 2.5% 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
5
Values in Thousands Values in 10^ -4
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Year M inim um M axim um M ean M
Std D eviation Variance Skew ness Kurtosis 1990 8511.8 100122.1 21261.6 20106.1 4907.0 24078210.0 2.343 16.888 2004 8085.4 79178.2 17999.0 19580.3 3780.6 14292880.0 2.006 14.693
1990
M ean = 21261.55 X <=33232.12 97.5% X <=14650.43 2.5%
0.0 0.2 0.4 0.6 0.8 1.0 1.2 28 55 83 110
Values in Thousands Values in 10^ -4 2004
M ean = 17998.98 X <=26813.37 97.5% X <=12535.62 2.5%
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 20 40 60 80
Values in Thousands Values in 10^ -4
Tier 1: 1990: 21.2 Tg CO2eq 2004: 18.0 Tg CO2eq std: 45% (50%)
2O emissions
Monte Carlo analysis: 1990: 21.3 Tg CO2eq 2004: 18.0 Tg CO2eq std: 46.2 % (1990); 42.0% (2004) 1990 2004
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Tier 1: Relative change: -1.5% std: 2.0%
2O emissions
Monte Carlo analysis: Relative change: -1.5% std: 3.4%
Mean = -3262.571 X <=-11899.73 2.5% X <=3144.66 97.5%
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
20
Values in Thousands Values in 10^ -4
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Y e a r M in im u m M a x im u m M e a n M
e S td D e via tio n V a ria n ce S ke w n e ss K u rto sis 1 9 9 5 6 4 .3 5 7 1 1 5 3 4 .3 6 8 7 3 4 .5 2 8 6 9 1 4 .6 1 1 9 2 .5 4 9 3 8 4 7 4 1 1 .1 .2 8 9 2 1 7 9 2 .4 6 6 6 8 1 2 4 9 4 1 .1 6 6 3 5 8 7 .0 1 3 2 2 5 1 .6 4 1 2 3 .9 5 3 1 6 .4 5 3 8 1 1 4 3
.0 1 2 5 4 7 4 2 .9 9 3 7 8 9
1 9 9 5
M ea n = 8 73 4.5 2 8 X < = 1 6 2 1 .54 9 7.5% X < = 7 1 5 1 .33 2 .5 %
.0 .5 1 .0 1 .5 2 .0 2 .5 3 .0 3 .5 4 .0 4 .5 6 8 1 1 2
V a lu e s in T h
sa n d s Values in 10^ -4 2 4
M e an = 22 51 .6 41 X < = 2 86 8.0 4 97 .5% X < = 1 62 9.7 2.5%
.0 .2 .4 .6 .8 1 .0 1 .2 1 .4 .5 1 .4 2 .3 3 .1 4 .0
V a lu e s in T h
sa n d s Values in 10^ -3
Tier 1: 1995: 8.25 Tg CO2eq 2004: 2.24 Tg CO2eq std: 28% (50% Jos)
Monte Carlo analysis 1995: 8.7 Tg CO2eq 2004: 2.25 Tg CO2eq std: 21.1 % (1990); 28.1% (2004) 1995 2004
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Tier 1: Relative change: -2.8% std: 0.4 %
Monte Carlo analysis: Relative change: -3.0% std: 0.9%
Mean = -6482.887 X <=-4778.3 97.5% X <=-8473.99 2.5% 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Values in Thousands Values in 10^ -4
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Re sulting unc e rta intie s a re in the sa me
the T I E R-1 F
inve nto ry, a c c o unting fo r c o rre la tio ns ha s a la rg e r impa c t o n the re sults fo r the tre nd tha n o n the unc e rta inty in the to ta l GHG e missio ns Ac c o unting fo r c o rre la tio ns a nd a symme tric a l distrib utio ns (in ste a d o f using NL T I E R 1 a ssumptio ns) did no t le a d to ra dic a lly diffe re nt unc e rta inty in to ta l GHG e missio ns.
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Unc e rta inty a ssumptio ns in the NL a re we ll in the ra ng e o f E uro pe a n Studie s Re sulting unc e rta inty in the to ta l Dutc h GHG e missio ns is in the lo we r ra ng e →sha re o f CO 2 e missio ns is hig he r → CO 2 e missio n fa c to rs a re re la tive ly we ll unde rsto o d a nd mo nito re d → lo we r unc e rta intie s
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Regression Sensitivity for TOTAL EMISSIONS NETHERLANDS 2...
Std b Coefficients
Gaseous Fuels / RISK/C88 .095 Coke Oven and BF gas / RIS.../F30 .106 NE / NO/K62 .108 NCV / RISK/D38 .114 NCV / RISK/F38 .114 Diesel Oil / NO/C195 .115 Volatized N from fertilize.../F103 .116
.134 NA / RISK/J71 .154 N from fertilizers, animal.../D104 .165
.168 Nitrogen input from applic.../F94 .17 1 Managed Waste Disposal .../H8 .185
.332 Nitrogen input from manure.../F95 .334 N from fertilizers, animal.../F104 .654
0.25 0.5 0.75 1
Nitrogen Leaching agr. soils IEF N2O Nitrogen input from manure applied to soils IEF N2O Nitric Acid Production IEF N2O Managed Waste Disposal on Land; CH4 emission Nitrogen input from synthetic fertilizers, IEF N2O Manure management; Cattle IEF CH4 Nitrogen Leaching agr. soils, AD Indirect N2O from NO2 comb.& ind. processes EF N2O Enteric Fermentation; 1. Cattle IEF CH4 Volatized N from fertilizers IEF N2O Road Transportation; Diesel Oil, aggregate AD Petroleum Refining; Liquid Fuels; aggregate AD Petroleum Refining; Liquid Fuels; IEF CO2 Coke Oven and BF gas IEF CO2
Energy Industries gaseous fuels IEF CO2
Ranking of uncertain inputs according to their contribution in t Ranking of uncertain inputs according to their contribution in the variance he variance ( (Regression sensitivity base case no LUFC total 2004 GHG emission
Regression sensitivity base case no LUFC total 2004 GHG emission)
)
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Regression Sensitivity for Total Netherlands (CO2-eq.)/F94
Std b Coefficients
L
.117
N from fertilizers, animal.../D104 .144
.146 Nitrogen input from applic.../F94 .149 1 Managed Waste Disposal .../H8 .16
Nitrogen in
N from fe
Nitrogen
Nitrogen input from manure.../F95 .29 1 Managed
N from fertilizers, animal.../F104 .551 N from fe
0.25 0.5
Nitrogen Leaching agr. soils IEF N2O, 1990 Nitrogen Leaching agr. soils IEF N2O 2004 Managed Waste Disposal on Land; CH4 emission 1990 Managed Waste Disposal on Land; CH4 emission 2004 Nitrogen input from manure applied to soils IEF N2O 2004 Nitrogen input from manure applied to soils IEF N2O, 1990 Nitrogen input from synthetic fertilizers, IEF N2O 1990 Nitrogen input from synthetic fertilizers, IEF N2O 2004 Nitrogen Leaching agr. soils AD 1990 HFC-23, Effective EF 1990 Manure management; Cattle IEF CH4 1990 Manure management; Cattle IEF CH4 2004 Nitrogen Leaching agr. soils, AD 2004 Enteric Fermentation, Cattle; IEF CH4 1990 Enteric Fermentation, Cattle; IEF CH4 2004 Petroleum Refining; Liquid Fuels; aggregate AD
Regression sensitivity base case no LUFC, trend Regression sensitivity base case no LUFC, trend
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
F ro m the dia g no stic dia g ra m: fo r the unc e rta inty in to ta l GHG e missio n impro ve me nts in o ur kno wle dg e o f the e missio n fa c to rs fo r c a te g o rie s indire c t N2O e missio ns fro m a g ric ultura l so ils, dire c t N2O e missio ns fro m a g ric ultura l so ils, indire c t N2O fro m c o mb ustio n a nd industria l pro c e sse s a nd e missio ns fro m ma nure ma na g e me nt mig ht b e g ive n the hig he st prio rity. I nspe c tio n o f the pe dig re e ta b le re ve a ls tha t the ma in pro b le m in the kno wle dg e b a se is in va lida tio n a nd e mpiric a l b a sis.
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
the T ie r 1 a sse ssme nt c o uld a lso b e impro ve d b y:
ie r 1 unc e rta inty inputs fo r la ndfills;
ie r 1 unc e rta inty o f a c tivity da ta fo r fue l c o mb ustio n in the c o mme rc ia l se c to r a nd
ie r 1 unc e rta inty inputs fo r indire c t N2O e missio ns fro m a g ric ultura l so urc e s
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
F
do e s no t c ha ng e sub sta ntia lly a nd the sha re o f CO 2 a nd no n-CO 2 g a se s is no t sub sta ntia lly diffe re nt fro m 2004, it se e ms justifie d to use T ie r 1 a s ma in me tho d fo r unc e rta inty a na lysis in the NI R. Ho we ve r, b e c a use o f o ng o ing e missio n re duc tio n e ffo rts a nd c ha ng e s o ve r time in the fue l mix a s we ll a s in the sha re s o f no n-CO 2 g re e nho use g a se s, we re c o mme nd re pe a ting the Mo nte Ca rlo a na lysis re g ula rly (e ve ry 4 ye a rs) a s pa rt o f the QA/ QC pro c e dure s.
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
200 400 600 800 2002 2003 2004 2005 Year ΣNEE (gC m-2 a-1)
Haarweg Cabauw Horstermeer Leliestad Fochtelooerveen Haastrecht Oukoop Stein
(c)
Jacobs et al., WUR, 2007 ME1
Flux tower network:
Uptake 0.9 ton C/ha/yr
Source 2.2 ton C/ha/yr
grasslands: Uptake 0.28 ton C/ha/yr NIR: drained organic soils Source 5.18 ton C/ha/yr x 223000 ha = Source of 4.2 Mton C Need for regionally specific and spatially explicit CO2 emission factors
Slide borrowed from Pavel Kabat WUR
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
Prior Post 2002 Prior Post 2002
1020 kT CH4 1350 kT CH4 Atmospheric verification suggests substantially higher CH4 emissions than reported:
1641kT CH4 for yrs 2000-2006
Vermeulen, ECN, in prep ME2
Bottom-up 1995 Top-down 2002
Slide borrowed from Pavel Kabat WUR
Copernicus Institute Sustainable Development and Innovation http://www.chem.uu.nl/nws/www/publica/publicaties2006/E2006-58.pdf
http:/ / w w w .chem . http:/ / w w w .chem .uu uu. .nl nl/ / nw s nw s/ w w w / / w w w / publica publica / publicaties2 0 0 6 / E2 0 0 6 / publicaties2 0 0 6 / E2 0 0 6 -
5 8 .pdf pdf