QUANTIFICATION OF METHANE SOURCE LOCATIONS AND EMISSIONS IN AN URBAN SETTING B51Q Characterizing, Modeling, and Extending Urban Metabolism II Eric Crosson,

QUANTIFICATION OF METHANE SOURCE LOCATIONS AND EMISSIONS IN AN URBAN SETTINGB51Q Characterizing, Modeling, and Extending Urban Metabolism II

Eric Crosson, Picarro

2

Quantification of Methane Source Locations and emission in an urban setting

• Contributors:– NOAA – Colm Sweeney Jocelyn Turnbull– Boston University – Nathan Phillips, Lucy Hutyra– Gas Safety Inc. - Robert Ackley– Picarro – Sze Tan, Chris Rella– Purdue University - Paul Shepson, Maria Obiminda Cambaliza– NIST - James Whetstone, Tony Bova, Kuldeep R. Prasad – NASA- Richard Koyler– Penn State University – Ken Davis, Scott Richardson

3

Motivation

• The identification and quantification of greenhouse gas emissions from urban centers are becoming of more interest.

• Recent measurements indicate that urban emissions are a significant source of Methane (CH4) and in fact may be substantially higher than current inventory estimates.

• As such urban emissions could contribute 7-15% to the global anthropogenic budget of methane*.

* Wunch, D., P.O. Wennberg, G.C. Toon, G. Keppel-Aleks, and Y.G. Yavin, Emissions of Greenhouse Gases from a North American Megacity, Geophysical Research Letters, Vol. 36, L15810, doi:10.1029/2009GL)39825, 2009.

4

The Problem: Quantifying CH4 Emissions using a Network of Analyzers, Establishing priors.

Tall or Cell Phone Towers

GHG emissions extracted usingWRF-CHEM inversion models

• For CO2- use inventories such as Vulcan as a starting point.

• For CH4- no such starting point exist.

5

• Develop methodologies to identify methane source locations and gather information on methane flux signals to use as priors.– Total methane emissions.– Partition methane emissions both by source and

geographically within an urban center.– Develop methods that enable fast and accurate data

collection and “real time” processing.

Objectives of this effort

6

Measurements

Real-timeMet data

AtmosphericModels

Oil & Gasprocessing

10

kg/m

in

EmissionsMaps

Taking a “snap shot”: Speed, specificity, and real-time feedback are the keys to success

A powerful combination of new technologies is enabling scientist to generate high quality data at rates never before possible.

auto

flight

7

For example: Brisbane, CA (methane map)

Brisbane, CA

San Francisco

Brisbane

5 miles

8

Speed is “everything”: Getting a “snap shot”, lot of area to cover in precious little time

0 10 20 30 40 50 600

5

10

15

20

Meth

ane C

oncentr

aio

n (

ppm

)

Time (minutes)

Brisbane, CA

Bay Area: 21,442 miles of road**Metropolitan Transportation Commission (2003)

9

Accuracy- What is the source?

PetroleumFacility

UncultivatedAreas

Brackishwater

Little or no wind

Brisbane, CA

10

Methane Signals Can be Confusing

Natural Gas Vehicles

Inefficient Vehicles

Storm Drains

Natural Gas Leaks

Sewer Systems

Petroleum Facilities

Landfills

11

Specificity: isotopic carbon-13 CH4 measurements

How can one help to identify the source?

0.0 0.1 0.2 0.3 0.4 0.5

-46

-44

-42

-40

-38

-36

0.0 0.1 0.2 0.3 0.4 0.5

-46

-44

-42

-40

-38

-36

- 45.7 per mil

PG&E Natural Gas (East Bay) = -37 per milAmbient = -45 per mil

3671 16th Street, San Francisco, CA

De

lta

13 C

H4 (

pe

r m

il)

1/CH4 Concentration (ppm-1)

- 37.2 per mil

-37.2 per mil

-45.7 per mil

12

Specificity: isotopic measurements while driving at the speed of traffic

Natural gasConc (ppm)

Delta (permil)

Unknown

Brisbane, CA

Brisbane, CA

13

Brisbane Site: looking from another angle

Brisbane, CA

PetroleumFacility

Delta = -53.3 per mil

UncultivatedAreaOld Landfill

14

Methane Capture System Leaks

Brisbane, CA

PetroleumFacility

Old Landfill

Delta = -53.3 per mil

Large Area Leak

15

Old landfill releasing methane in only two ways

1. Pipe connectors leaks.

2. Methane capture system is failing in one area of landfill.

Take away: In a very short time, able to identify specific problems that may or may not be easily rectified by landfill owner or city fathers.

16

Quantifying Methane Source Emissions

• Driving speeds of 20-30 mph

• Vacaville landfill

• Late evening wind conditions were very stable:– Speed 8.2 ±1.3

m/s– Direction 224.2 ±

5.0 deg

wind

• This was late evening with a very flat topography – estimated stability class is E or F (

17

Analyzing line integration data in real time to provide immediate feedback

• Post process data using more complex models at a later time

Source Location from model

18

Flux chamber measurements obtained using fast, high precision CH4 analyzers.

-26.

25

-22.

5-2

0

-17.

5

-16.

25

-13.

75

-12.

5-1

0

-8.7

5

-6.2

5-5

-2.5

-1.2

5 02.

53.

75 56.

258.

75 10

13.7

517

.520

05

101520253035

-2 4.00

45 feet

4 feet

Leak

Flu

x

• Using chamber measurements, one can obtain a lower limit on the total leak rate

• Found leaks ranging from 40 -300 sccf/day.

• As a guideline, the natural gas used in the “average” American home is 200 standard cubic ft per day.

Leak Rate > 41 cubic ft per day

Methane is coming out of the road.

19

Ready to tackle more complex systems?.....Methane Emissions in Indianapolis

Wind direction, Day 2

Methane Concentration levels are 3 times ambient 4 km from the landfill.

Wind direction, Day 1

Emissions from these locations on landfill

Little or no emissions from waste water treatment plant

Unknown source of methane

Natural Gas leak after gas station fire a few

days earlier

20

Summary

• Developed fast methodologies for identifying methane source locations and gathering information on methane flux signals to use as priors.

• For 2012:– Continue developing fast methodologies– Measure methane emission in several urban centers

• Indianapolis• Boston• San Francisco / Bay Area• Paris