wrap regional modeling center uc riverside/environ presented at rpo national workgroup meeting

WRAP Regional Modeling CenterUC Riverside/ENVIRON

Presented atRPO National Workgroup Meeting

November 4 – 6, 2003St. Louis, Missouri

WRAP Wind Blown Fugitive Dust and Ammonia Emissions Updates

Project Team

Fugitive Dust• Gerard Mansell; ENVIRON• Martinus Wolf, Paula Fields; ERG• Jack Gillies; DRI• Mohammad Omary; UCR• Bill Barnard; MACTEC Engr. & Consulting• Michael Uhl; DAQM, Clark County, NVAmmonia• Mark Chitjian; UCR• Gerard Mansell; ENVIRON

Outline

• Project Background & Overview• Literature Review• Estimation Methodology• Agricultural Considerations• Data Sources• Summary of Assumptions• Program Implementation• Results• Recommendations

Background and Overview of Project

• Develop General Methodology to Facilitate Future Revisions and Control Strategy Development

• Develop Integrated SMOKE Processing Modules for PM10 and PM2.5 Emissions Modeling

• Develop PM10 and PM2.5 Emission Inventory Applicable to the Western Region

Overview of Technical Approach

• Categorize Vacant Land Types• Identify Wind Tunnel Emission Factors• Develop Meteorological Data• Develop Threshold Wind Velocities, Wind Events,

Precipitation Events• Apply Emission Factors to Vacant Land Categories

Literature Review

• Portable field wind tunnels have been used to investigate particle entrainment thresholds, emission potentials, and transport of sediment by wind.

• Major contributions of information on: – thresholds from Gillette et al. (1980), Gillette et al.

(1982), Gillette (1988), Nickling and Gillies (1989);– emission fluxes from Nickling and Gillies (1989),

James et al. (2001), Columbia Plateau PM10 Program (CP3), Houser and Nickling (2001).

• Key information has also come from dust emission modeling (e.g., Alfaro et al., 2003) and desert soil characterization studies (e.g., Chatenet et al., 1996).

Wind Tunnel Study Results: Thresholds

u*t = 0.31e7.44x(Zo)

R2 = 0.60

u*t = 0.30e7.22x(Zo)

0

0.5

1

1.5

2

2.5

3

0.00001 0.0001 0.001 0.01 0.1 1

zo (cm)

u *t (

m s-1

)

wind tunnel data Marticorena et al. 1997Expon. (wind tunnel data) Expon. (Marticorena et al. 1997)

Comparison between modeled relationship of threshold friction velocity and aerodynamic roughness length and wind tunnel data.

*

*(Gillette et al., 1980; Gillette et al., 1982; Gillette, 1988; Nickling & Gillies, 1989)

Wind Tunnel Study Results: Emissions

FFSF = 2.45x10-6 (u*)

3.97

FSF = 9.33x10-7 (u*)

2.44

MSF = 1.243x10-7(u*)

2.64

CSF = 1.24x10-7 (u*)

3.44

0.000000001

0.00000001

0.0000001

0.000001

0.00001

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Friction Velocity (m s-1)

Em

issi

on F

lux

(F, g

cm

-2 s-1

)

FSSFSMSCSPower (FSS)Power (FS)Power (MS)Power (CS)

The emission flux as a function of friction velocity predicted by the Alfaro and Gomes (2001) model constrained by the four soil geometric mean diameter classes of Alfaro et al. (2003).

Wind Tunnel Study Results: Emissions as a function of texture

Sand(CS) Silty sand (MS) Sandy silt (FS) Silt (FSS)

Clayey silt

Silty clay

sand silt

clay

Clay

Sandy siltyclay

Clayey sandysilt

Clayey siltysand

Sand

y clay

Clay

ey sa

nd

50%

7525

7525

10

10

10

Sand(CS) Silty sand (MS) Sandy silt (FS) Silt (FSS)

Clayey silt

Silty clay

sand silt

clay

Clay

Sandy siltyclay

Clayey sandysilt

Clayey siltysand

Sand

y clay

Clay

ey sa

nd

50%

7525

7525

10

10

10

Sand(CS) Silty sand (MS) Sandy silt (FS) Silt (FSS)

Clayey silt

Silty clay

sand silt

clay

Clay

Sandy siltyclay

Clayey sandysilt

Clayey siltysand

Sand

y clay

Clay

ey sa

nd

50%

7525

7525

10

10

10

Relations between the soil types deduced from aggregate size distributions of various desert soils and soil textural categories (Chatenet et al. 1996). The “gray” highlighted textural classes indicate the 4 sediment types; the arrows indicate the pathways linking these types to the other textures. These can be linked to the North American soil texture triangle.

Wind Tunnel Study Results: Emissions

1E-10

1E-09

1E-08

1E-07

1E-06

10 100u* (cm/s)

Fv (g

/cm

2/s)

FS

Casa Grande

Mesa (Salt River )

Hayden

Ajo

Yuma (ag)

Glendale

Tucson (Sta Cruz)

Tucson (const.site)Mesa (ag)

CS

Yuma (scrubdesert)Yuma (distdesert)

Comparison between model relationship for FS and CS sizes and the wind tunnel data of Nickling and Gillies (1989). Ten (out of 13) sites have a dust production potential similar to the FS model and one site (Mesa agricultural) is closely aligned with the CS model (after Alfaro et al., 2003).

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

20 - 24.9 25 - 29.9 30 - 34.9 35 - 39.9 40 - 44.9 45 - 49.9 50 - 54.9

Emission Rates by Soil Group for Stable Soils Em

issi

on F

acto

r (to

n/ac

re/h

our)

Soil Group 1Soil Group 2

Soil Group 3Soil Group 4Soil Group 5

10-m Wind Speed (mph)

Emission Rates by Soil Group for Unstable Soils

0

0.005

0.01

0.015

0.02

0.025

0.03

20 - 24.9 25 - 29.9 30 - 34.9 35 - 39.9 40 - 44.9 45 - 49.9 50 - 54.9

10-m Wind Speed (mph)

Emis

sion

Fac

tor (

ton/

acre

/hou

r)

Soil Group 1Soil Group 2

Soil Group 3Soil Group 4Soil Group 5

Agricultural Considerations

• Non-climatic factors significantly decrease soil loss from agricultural lands

• Similar approach to CARB, 1997• Five “adjustment” factors simulate these effects:

– Bare soil within fields – Bare borders surrounding fields– Long-term irrigation– Crop canopy cover– Post-harvest vegetative cover (residue)

0.000

0.100

0.200

0.300

0.400

0.500

0.600

0.700

0.800

0.900

1.000

0 10 20 30 40 50 60 70 80 90 100

% Cover

Fact

or V

alue Canopy Cover Factor

Residue Cover Factor

Canopy Cover and Residue Cover Adjustment Factor

Agricultural Adjustment Factor Development

• New regional data collected for WRAP project:– Crop calendars with growth curves from Revised

Universal Soil Loss Equation (RUSLE2) model– Residues remaining after harvest due to

conservation tillage practices from Purdue’s Conservation Technology Information Center (CTIC)

– Irrigation events from crop budget databases • Factors applied by county/crop type, crop

management zones (CMZs)

Data Sources

• Land Use/Land Cover (LULC)– Biogenic Emission Landcover Database (BELD3)– North American Land Cover Characteristics – National Land Cover Database (NLCD)

• Soils Characteristics– State Soil Geographic Database (STATSGO)– Soil Landscape of Canada (SLC_V2)– International Soil Reference and Information Centre

• Meteorological Data– 1996 MM5 36-km (Wind Velocity, Precipitation,

Snow/Ice, Soil Temperature)

Land Use/Land Cover Data

• BELD3 LULC Data

Summary Total Area (Acres) % % excluding waterUrban 6,781,771 0.26% 0.34%Agriculture 531,231,552 20.54% 26.35%Shrub/grassland 720,022,464 27.84% 35.71%Forest 741,902,639 28.69% 36.80%Barren 5,801,931 0.22% 0.29%Wetlands 681,383 0.03% 0.03%Tundra 9,096,875 0.35% 0.45%Snow&Ice 603,210 0.02% 0.03%Water 569,829,853 22.04%Total 2,585,951,680 100.00%

Total excluding water 2,016,121,827 100.00%

Meteorological Data

• 1996 MM5– 1996 Annual, hourly, gridded meteorology– 36-km horizontal resolution– 10-m wind speeds– Precipitation rates– Snow/ice cover flag– Soil temperature

Data Compilation for Land Use and Soil Types

• Land use and soil texture aggregated to 12-km resolution

• Major land use categories– Urban– Agricultural– Shrub and grasslands– Forest– Barren and Desert

• Land use fractions from 1-km data retained as percentages

• Dominate soil texture at 12-km resolution

Soil Texture Categorization

• Standard soil types mapped to 5 major types for dust calculations– Silty Sand and Clay– Sandy Silt– Loam– Sand– Silt

STATSGO Soil Texture

Soil Texture Code

Soil Group Code

No Data 0 0Sand 1 4Loamy Sand 2 4Sandy Loam 3 2Silt Loam 4 1Silt 5 5Loam 6 3Sandy Clay Loam 7 2Silty Clay Loam 8 5Clay Loam 9 3Sandy Clay 10 2Silty Clay 11 5Clay 12 1

Vacant Land Stability

• Windblown dust emissions affected by soil stability• Stability determination based on land types• Urban lands may be stable or unstable

LULC Category Stability Urban Stable/Unstable (see below) Agricultural -- Shrubland Stable Grassland Stable Mixed Shrub/Grassland Stable Forest Stable Barren Unstable Desert Unstable

Urban Land Stability

• Urban lands divided into core and boundary areas• Core = 92.67%; Boundary = 8.33% • Core urban areas:

– 8 % unstable– 92% stable

• Boundary urban areas:– 30% unstable– 70% stable

Reservoir Characteristics

• Reservoirs characteristics based on stability– Stable = limited– Unstable = unlimited

• Stable reservoirs are depleted within 1 hour• Unstable reservoirs are depleted within 10 hours• Reservoirs require 24 hours to recharge

Precipitation and Freeze Events

• No dust emissions during rain events• Rainfall from MM5 at 36-km resolution• No dust emissions if snow/ice cover present• Dust emissions re-initiated:

– 72 hours after rain– 72 hours after snow/ice meltdown– 12 hours after thaw

Vegetative Cover Adjustments

• Vegetation cover reduces dust emissions• Methodology developed for bare soil• Emissions reduction factors developed from White

(2000)• Vegetation density based on land use types

Vegetative Cover Adjustments

LULC Category Vegetation Cover % Reduction Factor Urban 55(stable)/0(unstable) 0.07/1.0 Shrubland 11 0.70 Grassland 23 0.19 Mixed Shrub/Grassland 17 Forest 55 0.07 Barren 0 1.0 Desert 0 1.0

Summary of Assumptions

• Threshold velocity = 20 mph• Vacant land stability• Urban lands• Dust reservoirs• Reservoir depletion and recharge times• Precipitation, snow and freeze events• Vegetation density

Program Implementation

• Daily/Hourly Meteorological Data• State/County, Crop Management Zone, and Soil

Type, For Each 12km Cell.• Area fractions For Each 12km Cell, and Land Use

For Each Area Fraction.• Agricultural Adjustment Data• Emission Rates by Soil and Wind Speed Categories

Agricultural Input Data

• County and CMZs for 12-km grid cells• Crop area percentages for 12-km cells• Barren, border and crop fractions for BELD3 crops• Long term irrigation factors by soil type• Irrigation fraction by county and crop• Tillage fractions by crop and county• Planting and harvesting dates by crop (crop

calendars)• Crop canopy cover by crop (growth curves)• Residue cover

Summary of Annual PM10 Emissions

Annual PM10 by Landuse

urbanAg_nonadjustAgdesert,barren landforestshrubgrasslandshrub/grassland

PM10 Dust Emissions by Month

Total Dust Emission

00

200000

400000

600000

800000

1000000

1200000

1400000

1600000

1 2 3 4 5 6 7 8 9 10 11 12

Month

(tons

)

Monthly PM10 Emissions by Landuse Type

Total PM10 (tons)

0

200000

400000

600000

800000

1000000

1200000

1400000

1600000

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

Month

tons

water/welandshrub/grasslandgrasslandshrubforestdesert,barren landAgAg_nonadjusturban

Monthly PM10 Emissions by Crop Type

Monthly PM10 by crops

0

50000

100000

150000

200000

250000

300000

350000

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

Month

PM

10(to

ns)

WheatTobaccoSoybeansSorghumRyeRicePotatoesPeanutsPastureOatsMiscHayGrassCottonCornBarleyAlfal

Annual PM10 and PM2.5

Monthly PM10 Emissions

Monthly PM10 Emissions

Monthly PM10 Emissions

Monthly PM10 Emissions

Monthly PM10 Emissions

Monthly PM10 Emissions

Sensitivity Simulation

• Evaluate impact of threshold velocity and reservoir assumptions

• Extend emissions factor relations to lower wind speeds

• Relax reservoir recharge assumptions:– 6 hours between wind events– 24 hours after rain events– 24 hours after snow/ice meltdown– 6 hours after thaw

Recommendations

• Methodology review and refinement• Current, detailed data to characterize vacant lands• Methodology validation with small-scale, high

resolution domain• Comparison with other methodologies• Identification and evaluation of additional wind

tunnel studies• Application to other domains, years

WRAP Ammonia Emissions Updates

Activity Data

EmissionFactors

NLCDCensus Data

Meteorological Data

Soils Data

AllocationAssociation

Table DomainDefinition

Annual Inventory

-by source -by county

ReportsSummaries

QA/QC

Gridded Inventory

-total NH3 -by grid cell

EmissionEstimation

Spatial andTemporalAllocation

ArcGisAmmoniaEmission

Model

Model Flow

• Designed to :– treat relationship of ammonia emissions to environmental variables– incorporate high resolution spatial data (LU/LC databases; population

density; geo-coded point sources)– incorporate detailed source classification schemes

• Based on Arc/INFO GIS• Flexible input data sources/formats; easily modified input tabular data

(activity data, EF, spatial surrogate relationships, domain definition)• Efficient processing capabilities • Modular• Intuitive user interface

GIS-Based Emissions Model Development

• Livestock - dairy, beef, pigs, poultry, sheep, horses

• Fertilizer

• Soils – highly uncertain, may at time act as source or sink, potentially very large source

Focus on Largest Sources

Minor Sources

•Mobile - on-road and off-road•Domestic - respiration, perspiration, cigarettes, pets, etc.•Landfills•Composting•Industrial •Wildfire/Ag fire

Spatial Allocation

• Point Sources – As many sources as possibleLandfills Confined Feeding OperationIndustrial Large Poultry FacilitiesCompost Dairies

• National Land Cover Data (NLCD)• Census Coverages – Domestic Sources• Environmental Parameters – i.e., Soil pH

Temporal Allocation

• Empirically Based Seasonal and Diurnal Profiles

• Temporally Varying Environmental ParametersWind SpeedAir Temperature Soil Surface Temperature

NLCD

• 30 meter resolution• 21 Land Cover Classification Categories• Based on Landsat Thematic Mapper Data• Developed by Multi-Resolution Land Characterization

Consortium• USGS• EPA• Forest Service• NOAA

NLCD Classification Codes

11 Open Water 12 Perennial Ice/Snow21 Low Intensity Residential 22 High Intensity Residential 23 Commercial/Industrial/Transportation31 Bare Rock/Sand/Clay32 Quarries/Strip Mines/Gravel Pits33 Transitional41 Deciduous Forest42 Evergreen Forest43 Mixed Forest

51 Shrubland61 Orchards/Vineyards/Other71 Grasslands/Herbaceous81 Pasture/Hay82 Row Crops83 Small Grains84 Fallow85 Urban/Recreational Grasses91 Woody Wetlands92 Emergent Herbaceous Wetlands

WRAP NH3 Model Main Menu

Emission Calculation Menu

Annual Gridded Domestic NH3 Emissions

Literature Review

Eleven Relevant Recently Published Papers• Temporal Allocation – Sakurai and Fujita, 2002; vanHove

et al., 2002; Pinder at al., 2003; Roelle and Aneja, 2001;De Visscher et al., 2002; Anderson et al., 2003

• Livestock Emission Factors – Jarvis and Ledgard, 2002; Battye et al., 2003; Pinder at al., 2003Keener et al., 2001; Doorn et al., 2002; De Visscher et al., 2002

• Mobile Emission Factors – Durbin et al 2002• Soil Emission Factors – Battye et al., 2003; Roelle and

Aneja 2001

Schedule

• Draft 1996 NH3 modeling inventory by December 2003

• Next Step: 2002 NH3 inventory