Transcript Document 7294755

Model
Description
Model Overview
Written in Java
 Open Source / cross platform

 71

source files
After encountering memory issues:
○ Switched to a pre-compiled version of the program
(Excelsior JET package)

4 other programmers:
○ Roger Schlachter, Chris Schmitt, Andy Kuhn, and
Andy Korth
 Assistance from several JAVA library developers
Model Description

Overall Data Requirements:
 Site Properties
 Cover Properties
1. Site Properties
 Latitude
Also available through SWIS database
 Longitude
○ Needed for climate simulation
 Size of landfill – total waste footprint (acres)
SWIS Search

SWIS database
2. Cover Properties Panel
• Up to 10 different covers allowed in each model run
Cover Panel: Cover Details

Cover type:
 Daily, Intermediate, or Final

Coverage %
 Area coverage at the site for
this particular cover type
Default Boundary Conditions
Cover Type
Lower
Temperature
Boundary
Condition
(oC)
Lower Methane
Boundary
Condition
(% vol)
Lower Oxygen
Boundary
Condition
(% vol)
Daily
25
1(*)
5
Intermediate
35
45
1
Final
40
55
0
* Field data actually significantly lower (3000 ppm Marina and 3 ppm Scholl)
Cover Panel: Cover Details
Cover Properties scroll bars
 Organic Matter content
 Gas Recovery
 Vegetation Presence
Organic Matter Percentage

Impacts hydraulic
conductivity and water
holding capacity
(Benjamin et al., 2008)

Low to High
corresponds to
0 to 5%
Cover Panel: Gas Recovery

Gas Recovery Percentage
 This is NOT the recovery “efficiency”
 Indicates the spatial coverage of the gas recovery system
for that particular cover
Cover Panel: Gas Recovery Coverage %
To LFG
recovery
system
33% Gas Recovery Coverage
To LFG
recovery
system
100% Gas Recovery Coverage
To LFG
recovery
system
50% Gas Recovery Coverage
Vegetation Percentage
Estimated on surface coverage of vegetation
 Main impact:

 Decreases incident solar radiation to the soil
surface
○ Reduces heating and soil surface evaporation
Cover Panel :
Custom Cover Editor
Cover Editor :
Once layer highlighted:
1. Edit layer material – 33 different materials
•
12 Textural soil types (USDA soil classifications)
•
SILTY CLAY, CLAY, SILTY CLAY LOAM, SILT, CLAY
LOAM, SILTY LOAM, SANDY CLAY, LOAM, SANDY
CLAY LOAM, SANDY LOAM, LOAMY SAND, SAND
Cover Editor :
Once layer highlighted:
1. Edit layer material
•
12 Soil types (USDA soil classifications)
•
•
SILTY CLAY, CLAY, SILTY CLAY LOAM, SILT, CLAY LOAM, SILTY LOAM, SANDY
CLAY, LOAM, SANDY CLAY LOAM, SANDY LOAM, LOAMY SAND, SAND
21 Alternative Daily Covers and other materials
•
ADC Materials (e.g. tarps, foams, shredded tires, compost)
•
Other
•
Geomembranes (HDPE, LDPE, EDPM); Geotextiles
•
Rocks, Pebbles (pea gravel)
•
Contaminated soils
•
Foundry sands
•
Wood Chips
Cover Editor :
Once layer highlighted:
1. Edit layer material
2. Specify thickness of the materials
3. Layers arranged from top to bottom
Cover Editor:
Once layer highlighted:
1. Edit layer material
2. Specify thickness of the materials
3. Layers arranged from top to bottom
Default Final Covers


If the cover type of
final cover is
selected
5 default final cover
designs are in a pull
down combo box
Default final cover designs
Layer
CCR Title 27
Clay (without
geomembrane)
Geosynthetic
Cover (with
geomembrane)
Water Balance
(Vegetation
Surface)
Water Balance
(rock armored)
1
Loam
(12 inches)
Loam
(12 inches)
Loam
(12 inches)
Loam
(12 inches)
Rocks/Boulders
(6 inches)
2
Clay
(12 inches)
Clay
(40 inches)
HDPE
geomembrane
(1 inch*)
Silty Clay Loam
(36 inches)
Loam
(12 inches)
3
Silty Clay Loam
(24 inches)
Silty Clay Loam
(12 inches)
Silty Clay Loam
(24 inches)
Vegetation
50%
50%
50%
Silty Clay Loam
(36 inches)
50%
* Minimum layer size = 1 inch. HDPE is modeled as 40 micron HDPE with sand
0%
3. Simulated Weather

CALMIM uses 3 previously validated
simulation models
○ GlobalTempSIM
○ GlobalRainSIM
○ SolarCalc
Max (C)
Min (C)
Max
40.00
Min
Temperature (C)
Comparisons of
Weather Simulator
Oroville,Oroville,
CA CA
(2009)
50.00
30.00
20.00
10.00
0.00
0
50
100
-10.00
150
200
250
300
350
400
Day of Year (2009)
Fresno,Freesno,
CA (2009)
CA
Max (C)
50.00
Min (C)
Max
40.00
Temperature (C)
Min
30.00
20.00
10.00
0.00
0
50
100
-10.00
150
200
250
300
350
400
Day of Year (2009)
Long Beach,
Long Beach,
CACA (2009)
Temperature (C)
Max (C)
45.00
Min (C)
40.00
Max
35.00
Min
30.00
25.00
20.00
15.00
10.00
5.00
0.00
0
50
100
150
200
Day of Year (2009)
Weather Data from Weather Underground
250
300
350
400
Comparisons of
Weather Simulator
900
800
700
600
500
400
300
200
100
0
Oroville, CA (2009)
Global RainSIM
2009
2006
0
50
100
450
Global RainSIM
400
2009
350
2006
150
200
250
300
350
400
Fresno, CA (2009)
300
250
200
150
100
50
0
0
50
100
150
200
250
300
350
400
350
400
Global RainSIM
2009
2005
600
Long Beach, CA (2009)
500
400
300
200
100
0
0
Weather Data from Weather Underground
50
100
150
200
250
300
Soil Temperature and Moisture Model

Uses STM2 model
(Spokas and Forcella, 2009)

Some modifications:
 Moisture boundary conditions –
○ Upper = evaporation/precipitation boundary
○ Lower = saturated or free drainage {saturated default condition}
 Temperature boundary conditions –
○ Upper temperature boundary

Air temperature (simulated) or fixed temperature
○ Lower temperature boundary:
 Function of cover type:
- Daily : Average annual air temperature
- Intermediate : 40 oC
- Final : 45 oC
Gas Modeling

Uses already developed 1-D gas transport
model (Campbell, 1985)
 Added empirical soil oxidation equations from
laboratory testing for methanotrophic methane
oxidation – specific for CA soils
 1-D Oxygen diffusion modeled
 1-D Methane diffusion modeled
○ With and without CH4 oxidation
Supporting Laboratory Studies for Methane
Oxidation Modeling
•A total of 2,112 soil incubations have been completed using Marina and
Scholl Canyon cover soils
•Temperature range of 0-70 C and moisture range of -15 bar to zero soil
moisture potential
•Incubators: Isothermal and diurnal fluctuations
•Detailed in published paper Spokas and Bogner, 2010.
Laboratory Methane Oxidation Testing
Relative CH4 Oxidation Rate
Incubation Temperature < 5oC
Incubation Temperature 5-40oC
1.0
1.0
1.0
0.8
0.8
Relative rate 
0.6
0.6
R  0.994
0.4
0.4
0.4
0.2
0.2
0.2
0.852
1 e
0.0
-2000
-1500
-1000
-500
0
0.0
-2000
-1500
0.6
-1000
-500
Soil Moisture Potential (kPa)
relative rate  1.05e
 temp- 27.6 
 0.5

 9.59 
0
0.0
-2000
-1500
2
R 2  0.981
•27 oC maximum from lab data
0.6
0.4
•400 ug CH4 /gsoil/day average
0.2
0.0
0
10
20
30
40
o
Temperature ( C)
50
60
-500
Soil Moisture Potential (kPa)
0.8
-10
-1000
•The soil moisture potential for oxidation
for 50% of activity for the two validation
sites  -600 kPa
1.2
1.0
0.8
 SMP  754.1 
-

 151.84 
2
Soil Moisture Potential (kPa)
Relative CH4 Oxidation Rate
Incubation Temperature >40oC
70
0
Advanced Settings
Beta Testing

Initial testing conducted via USDA SharePoint
server
○ Total # of registered users = 52

Due to file compatibility issues (EXE files) and
delay in getting individuals registered with
SharePoint moved to http://calmim.lmem.us
○ 55 downloads of versions up to 4.0
○ 83 downloads of version 4.2 (newest)
Beta Testing Results
10 users provided detailed comments
 Identified memory issues early on in the
Beta testing

 Java memory handling is poor; particularly
for finite difference modeling
 Reprogrammed calculation routines to avoid
new variables; still occasional issues
Field
Validation
Marina Landfill Comparison
15 cm (intermediate cover) and 50 cm depth (final cover)
25
Temperature (C)
20
15
10
5
Correlation
Pearson's
d-index
Model Efficiency
RMSE:
MAE:
0
1-Aug
20-Sep
0.937
0.877
0.941
0.756
1.362
1.180
9-Nov
Model
50 cm Measured
29-Dec
17-Feb
7-Apr
27-May
16-Jul
4-Sep
24-Oct
Date/Time (2007-2008)
24
Temperature (C)
23
22
21
20
19
18
17
16
Correlation
Pearson's
d-index
Model Efficiency
RMSE:
MAE:
15
21-Aug
23-Aug
0.957
0.915
0.933
0.677
1.864
1.534
25-Aug
Model
27-Aug
29-Aug
31-Aug
2-Sep
Date/Time (2007-2008)
4-Sep
15 CM Measured
6-Sep
8-Sep
10-Sep
Soil Moisture - MIC
•Variable thickness of wood chips across area
•Reduces evaporation losses and increase soil
moisture (0-8” thickness)
•Model with 2” wood chips
Volumetric Soil Moisture
0.3
0.25
without woodchips
with woodchips
0.2
0.15
0.1
0.05
0
0
50
100
150
200
Day of Year
250
300
350
Soil Moisture: MDC
Marina Daily Cover Comparisons
Field Measurement
(g m-2 day-1)
March 2007
0.209
August 2007
0.564
March 2008
10.249
August 2008
8.860
Overall Range
0.2 to 10.2
Different
locations
Comparing CALMIM Output
MDC
Model overestimates CH4
oxidation for Daily Covers
Rates of CH4 oxidation at various depths under 3 conditions:
•No Pre-incubation at field-collected moisture
•60-day pre-incubation at field-collected moisture
•60-day pre-incubation at field-capacity moisture (33 kPa)
Marina
Intermediate
Depth (cm)
Daily
Daily
(ug CH 4 g soil-1 day-1)
A. Initial Rate Š No Pre-incubation at field collected moisture contents
0-10 cm
10-20 cm
0.05 (0.02)
0.04 (0.08)
0.4 (0.2)
1.9 (0.4)
0.3 (0.1)
0.2 (0.1)
0.1 (0.2)
0.1 (0.1)
20-30 cm
30-40 cm
40-50 cm
50-60 cm
70-80 cm
#
2.5 (0.6)
171.3 (22)
211.2 (36)
#
2.8 (0.5)
2.6 (0.2)
0.5 (0.3)
1.4 (0.2)
0.4 (0.2)
#
-1
Final
no
Scholl Canyon
Intermediate
Final
0.2 (0.3)
0.2 (0.1)
pre-incub;
-
0.2 (0.1)
0.2 (0.3)
0.1 (0.1)
field moist.
-
oxid. range = 0.05 -211
-1
B. Pre-incubation with 50 ml l CH 4 and 200 ml l O2 at field collected moisture contents
0-10 cm
10-20 cm
0.4 (0.2)
1.8 (0.1)
0.1 (0.3)
1.9( 0.1)
3.6 (0.5)
2.8 (0.4)
1.7 (0.2)
3.6 (1.4)
20-30 cm
30-40 cm
40-50 cm
50-60 cm
70-80 cm
#
8.9 (0.4)
384.2 (10.3)
374.1 (7.1)
#
5.6 (10)
111.3 (12)
199.8 (14)
219.8 (28)
212.7 (23)
#
0.2 (0.1)
0.2 (0.1)
pre-incub;-
0.9 (0.2)
0.5 (0.1)
field
0.2 (0.5)
moist.
-
oxid. range = 0.1 - 384
C. Pre-incubation with 50 ml l -1 CH 4 and 200 ml l -1O2 (60 d) at field capacity moisture content (33 kPa)
0-10 cm
10-20 cm
20-30 cm
30-40 cm
40-50 cm
50-60 cm
70-80 cm
142.2 (33)
132.6 (20)
#
416.8 (16)
412.9 (13)
412.7 (15)
412.4 (23)
452.0 (12)
#
593.8 (31)
573.9 (14)
613.1 (14)
594.2 (16)
604.2 (15)
NS
644.2 (28)
112.4 (19)
112.1 (13)
#
211.4 (32)
212.4 (39)
pre-incub; -field
-
212.9 (22)
212.7 (18)
212.5 (11)
capacity
-
oxid. range = 112 -644
(averages of 6 replicates; SD in parentheses)
MDC: Cover Thickness
8”
3000 ppm CH4
at base of cover
12”
16”
Daily Cover
Waste
Surface Emissions
with Oxidation
(g m-2 day-1)
Surface Emissions
without Oxidation
(g m-2 day-1)
8”
3.49
7.64
12”
0.06
4.83
16”
0
3.52
Thickness
Field data  0.2 to 10.2 g/m2/day
MDC: Cover Thickness
annual emissions (g m-2 d-1)
450
400
350
300
y = 2090.3x -1.167
R² = 0.9989
250
200
150
100
50
0
0
5
10
15
thickness (in)
20
25
MDC:
Concentration at Base of Cover
12”
1000 ppm CH4
at base of cover
3000 ppm CH4
at base of cover
Daily Cover
10000 ppm CH4
at base of cover
Waste
Surface Emissions
with Oxidation
(g m-2 day-1)
Surface Emissions
without Oxidation
(g m-2 day-1)
1000 ppm
0
1.61
3000 ppm
0.06
4.83
10000 ppm
9.58
16.11
Concentration @
base of cover
MDC:
Concentration at Base of Cover
Surafce Emission (g/m2/day)
18
16
14
12
10
8
6
1000 4
ppm CH4
at base
2 of cover
0
0
12”
2000
3000 ppm CH4
at base of cover
Daily Cover
 C 
Linear relationsh ip : J  De 


x


10000 ppm CH4
at base of cover
4000
6000
8000
10000
12000
Waste
Base of Cover Concentration
(ppm)
Surface Emissions
with Oxidation
(g/m2/day)
Surface Emissions
without Oxidation
(g/m2/day)
1000 ppm
0
1.61
3000 ppm
0.06
4.83
10000 ppm
9.58
16.11
Concentration @
base of cover
Marina Intermediate Cover
[Cover placed late 2006]
Field Measurement % CH4 (v/v) at base
of cover
(g m-2 day-1)
March 2007
0.03
11.2
August 2007
53.17
40.7
March 2008
34.20
54.9
August 2008
237.81
54.4
Average Field Data Range
Estimation From13C Isotope Probe Samples
Marina Final Cover Comparisons
Field Measurement
(g m-2 day-1)
Overall Average
CALMIM
March 2007
0.00
August 2007
0.00
March 2008
0.01
August 2008
0.10
0.03
0 with oxidation
23 without oxidation
Scholl Canyon : Daily Cover
Field Measurement
(g m-2 day-1)
CALMIM
March 2007
0.003
August 2007
0.004
March 2008
0.008
August 2008
-0.001
<0.01
Scholl Canyon Intermediate Cover
Field Measurement
(g m-2 day-1)
CALMIM
March 2007
-0.006
August 2007
0.002
March 2008
0.013
August 2008
-0.003
<0.01
Scholl Canyon Final Cover
Field Measurement
(g m-2 day-1)
CALMIM
March 2007
0.006
August 2007
0.015
March 2008
0.019
August 2008
0.022
<0.01
Questions from Beta Testers

Why are the emission results for different landfill sizes (waste in
place) the same, if the same cover and cover type (e.g.
concentration profiles) are used?
 The assumptions for this model:
○ Diffusion is the dominant transport mechanism: Concentration gradient
controls gas transport
If concentrations are equal, then
emissions are equal
Advection requires
connectivity (e.g. soil
cracks) for gas flow.
 C 
J  De 


x


Questions from Beta Testers

Why are the emission results for different landfill sizes the same
(i.e. waste in place), if the same cover and concentration profiles
are used?
 The assumptions for this model:
○ Diffusion is the dominant transport mechanism
○ Concentration gradient controls gas transport
 Departure from first-order gas generation modeling
 Potentially an easier to measure field parameter (gas
concentration at base of cover) versus degradation
constants/WIP stats/composition data…
○ Greatly reduces uncertainty in the modeling for inventory purposes
Other comments from Beta Testers

Additional user guidance needed
○ Completing user manual

Output written to Excel compatible files
 Worksheet tabs with output as a function of date and depth of various
properties
 Created in “My Documents\CALMIM-DataOutput\SiteName\”

Final Modifications being Completed
○ Still reducing memory leaks.. working with the programmers of
the various libraries to improve memory performance
Summary and Conclusions:
Project has developed a new GHG Inventory Methodology for landfill
methane based on:
expansion and integration of existing field-validated modeling approaches for meteorology
and soil microclimate, including use of publicly-available climatic databases
site-specific cover soils and areas with gas recovery
new modeling for methane emissions inclusive of seasonal methane oxidation in cover soils
field validation over 2 annual cycles
supporting laboratory incubation studies for methane oxidation
Just the first step ; not the end of the road
Model available at:
http://calmim.lmem.us
The difficulty lies, not in the new ideas, but in escaping the old ones,
which ramify, for those brought up as most of us have been, into every corner of our minds
-John Maynard Keynes
Quoted in: K. Eric Drexler Engines of Creation: The Coming Era of Nanotechnology, Bantam, New York, 1987, p 231.
NEW (THIS PROJECT)
CO2
CH4
recovered
CH4
methanotrophs
gas
well
migration
OLD (PREVIOUS INVENTORIES)
methane oxidation
in aerobic zone
this
method:
including
site-specific
cover
materials,
seasonal
climate,
WITH field
validation
emission
methane production in
anaerobic zone: methanogens
with 10% default for oxidation and
national methane recovery
previous
methods:
IPCC national
inventory
methods;
US EPA
LandGEM;
GASSIM
Acknowledgments:
CEC, CIWMB, ARB
Special thanks to the field sampling crew:
Chad Rollofson, Martin duSaire, and Dean Peterson
Field Validation Sites:
Scholl Canyon Landfill (Los Angeles County Sanitation Districts)
Marina Landfill (Monterey Bay Regional Waste Management Authority)