monitoring needs & issues bill malm national park service may 15, 2008
TRANSCRIPT
Monitoring Needs & Issues
Bill MalmNational Park Service
May 15, 2008
What modifications in the national monitoring networks are
required to assess effects and apportion the many and varied chemical species to their many and varied emission sources?
NEEDS
• Assessments– Ecosystem health
• Total deposition estimates (wet and dry) of all sulfur and reactive nitrogen species –
• Critical loads
– Visibility– Health (other)
• Attribution of each species to its emission source (control measures)
Issues
• Time scale of sample collection.– 1 week deposition monitoring– Every third day for IMPROVE/STN– Semi-continuous?
• Accuracy with which some species are measured
• Some key species are not measured at all
What is reactive nitrogen?
• The term reactive nitrogen includes all biologically active, photochemically reactive, and radiatively active nitrogen compounds in the atmosphere and biosphere of the earth. (both reduced and oxidized forms)
Wet deposition patterns and
trends
• N deposition “hot spot” in northern Colorado Rockies– Current N deposition ~ 20x
natural levels
• Nitrogen deposition increasing at high elevation sites– Ammonium deposition
increasing faster than nitrate
NADP 2004 Annual Summary
Burns (2003)Niwot Saddle NADP site
Are Fires Contributing to the Increasing Wet Nitrogen Deposition?
Wet nitrate concentration deposition trends
Wet ammonium concentration deposition trends
20th percentile,insignificant change
20th percentile,significant change
80th percentile,insignificant change
80th percentile,significant change
• Rocky Mountain National Park (ROMO) is experiencing a number of deleterious effects due to atmospheric nitrogen and sulfur compounds. These effects include visibility degradation, changes in ecosystem function and surface water chemistry from atmospheric deposition, and human health concerns due to elevated ozone concentrations.
• The nitrogen compounds include both oxidized and reduced nitrogen. Emissions of nitrogen compounds need to be reduced to alleviate these deleterious effects. Various regulatory programs are underway to address emission reductions, many of which will be achieved from the most easily identified contributors to oxidized nitrogen related effects at the park.
STATEMENT OF THE PROBLEM
ROcky Mountain Atmospheric Nitrogen and Sulfur study (ROMANS)
• Increased haze reducing visibility • Low capacity to sequester atmospheric N
deposition• N enrichment and shifts in diatom communities
in alpine lakes• N enrichment in organic soil layer and
Engelmann spruce needles on eastern slope
Concerns about increasing reactive nitrogen in Rocky Mountain National Park (Typical of all high mountainous regions)
Why Ammonia?• Direct ecosystem effects• Response of a basic gas neutralizing acidity in
particles and gases. (neutralization of acidic sulfate aerosols – reaction with nitric acid vapor – reactions with organic salts)
• Response of PM formation can be dislocated from where ammonia reduction first took place.
• Ammonia deposition via cloud uptake and subsequent rain, dry deposition in the gas vs particle phase have vastly different time scales that leads to different lifetimes and particle response.
The urban/industrial-agricultural interface
NOx
HNO3 + NH3 NH4NO3(p)+hv
Organic nitrogen?
???
VOC
•Characterize the atmospheric concentrations of sulfur and nitrogen species in gaseous, particulate and aqueous phases (precipitation and clouds) along the east and west sides of the Continental Divide (Organic Nitrogen?)
– GAS: NH3, R-NH2, NOX(NO+NO2), NOY(HNO3, PAN, etc)
– PARTICLE: NH4, NO3, ORGANICS (reduced and oxidized)?
– WET (rain and clouds): NH4, NO3, ORGANICS (reduced and oxidized)?
•Identify the relative contributions to atmospheric sulfur and nitrogen species in RMNP from within and outside of the state of Colorado.
•Identify the relative contributions to atmospheric sulfur and nitrogen species in RMNP from emission sources along the Colorado Front Range versus other areas within Colorado.
•Identify the relative contributions to atmospheric sulfur and nitrogen species from mobile sources, agricultural activities, large and small point sources within the state of Colorado.
•Characterize the atmospheric concentrations of sulfur and nitrogen species in gaseous, particulate and aqueous phases (precipitation and clouds) along the east and west sides of the Continental Divide (Organic Nitrogen?)
– GAS: NH3, R-NH2, NOX(NO+NO2), NOY(HNO3, PAN, etc)
– PARTICLE: NH4, NO3, ORGANICS (reduced and oxidized)?
– WET (rain and clouds): NH4, NO3, ORGANICS (reduced and oxidized)?
•Identify the relative contributions to atmospheric sulfur and nitrogen species in RMNP from within and outside of the state of Colorado.
•Identify the relative contributions to atmospheric sulfur and nitrogen species in RMNP from emission sources along the Colorado Front Range versus other areas within Colorado.
•Identify the relative contributions to atmospheric sulfur and nitrogen species from mobile sources, agricultural activities, large and small point sources within the state of Colorado.
ROMANS OBJECTIVES
RoMANS measurement network
• Two measurement campaigns
– Mar/Apr 2006
– July/Aug 2006
• Spring and summer historically have highest deposition fluxes
• 4 site types– Continuous
to weekly monitoring
Spring overview
• Concentrations lower in mountains
• Gases dominate at eastern and western sites– Highest
ammonia at Brush in NE Colorado
• Particles dominate in mountains
Summer overview
• Concentrations higher than in spring
• Highest concentrations again east of RMNP
• Increasing N gas importance in mountains
wet ON20%
wet no320%
wet nh426%
hno37%
nh310%
no31%
nh43%
nox4%
ONGas ?9%
wet no322%
nh312%
no30%
hno39%
wet nh422%
nh42%
nox2%
wet ON ?21%
ON Gas ?10%
• Total N deposition was about twice (2) as high during the summer vs spring.
• About 45% of N deposition is not being measured in the current monitoring programs (NAPD & CASTNET).
• Deposition of N is about 2/3 wet (rain and snow) and 1/3 dry (particles and gases).
• Organic N may be about 30% of total
deposition and is not currently being measured.
SOME PRELIMINARY RESULTS
MONITORING NETWORKS
• IMPROVE• STN• CASTNET• NADP/AIRMON• MERCURY• NAAQS• Persistent Organic Pollutants (POPs) • Others?
IMPROVE (24 hr every third day)
• Dry Species (Current)– SO4 – NO3
– Total POM– Metals (soil and attribution)
• Dry Species (Missing)– NH3/NH4
– Reduced and oxidized organic nitrogen containing particulates– NOx (NO and NO2)– Oxidized organic gases (PAN - alkyl nitrates …)– Reduced organic gases (Aliphatic amines …..)
CASTNET (weekly)
• Dry Species (Current)– SO2/SO4 – HNO3/NO3
– NH4
• Dry Species (Missing)– NH3
– NOx (NO and NO2)– Oxidized organic gases (PAN - alkyl nitrates …)– Reduced organic gases (Aliphatic amines …..)– Reduced and oxidized organic nitrogen containing
particulates.
Continued
• Wet Species (Current)– SO4
– NO3
– NH4
• Wet Species (Missing)– Organic nitrogen
• Reduced• Oxidized• Biological/terrestrial
Accuracy/Uncertainty (IMPROVE)
• The species that are measured are done so with reasonable accuracy and precision
Accuracy/Uncertainty (CASTNET/NADP)
• SO2/SO4 measured reasonable well for both wet and dry
• Nitrogen is problematic across the board– Cut point is ill defined (coarse vs fine)– HNO3/NO3 split has large error– NH4 error (underestimated) may be on order
of 20-50%– NH4 and NO3 may be biological converted
over the course of a week.
What isn’t measured?
• NH3
• Organic nitrogen either in wet or dry (gas and particle phase) or its reduced, oxidized or biological forms– NO2, peroxyacetyl nitrate (PAN) and related
alkyl nitrates– Aliphatic amines– Proteins, amino acids, etc
How important is AON?
• 51 studies in North America have DON at 38±19% of TON (Wet)
• As much as 30% of particulate OC is nitrogen containing
• Gas % of TON ?
Wet DON
• Can measure TON• Can’t directly measure reduced, oxidized, or biological
ON – important to make these distinctions from an apportionment perspective because sources are distinctive
• Can use receptor type models to apportion wet DON if one has reliable chemical markers– Measure reduced OC markers such as amines and urea for
reduced DOC– Measure oxidized OC markers such as alkyl nitrates,
nitrophenols, and other nitroaromatic– Measure biological markers such as amino acids and peptides – Apply simple regression models or more sophisticated models
such as PMF and UNMIX.
Dry gas and particle ON
• Can measure both oxidized and reduced ON both in gas and particle phase using catalytic converters and collect in near real time.
Recommendations from EPA NAAQS review: deposition index
from NOX/SOX review
• We recommend monitoring a suite of reactive nitrogen species the sum of which is “Total Chemically Reactive Nitrogen” defined as the sum of all oxidized species except N2O and the sum of ammonia and ammonium.
• Total Chemically Reactive Nitrogen = NOy + NHx + ?
• Species Method NOy (total oxidized nitrogen) Reduction to NO followed by chemiluminescence NO3
- (particulate nitrate) Denuder/filter sampling followed by ion chromatography HNO3 (nitric acid vapor) Filter/denuder and followed by ion chromatography. NH3 (ammonia) Filter/denuder followed by colorimetry or ion chromatography NH4+ (ammonium) Denuder/filter followed by colorimetry or ion chromatography
Questions
• Is split between various species important– SO2/SO4, HNO3/NO3, NH3/NH4, etc or is total
sulfur, or total gas/particle phase reactive nitrogen adequate?
• For ecosystem response may not be so important?• For attribution and model assessment it is critical!
• Can defensible critical loads be set without a knowledge of total nitrogen?
• What sampling frequency/duration is acceptable – both in time and space?
Temporal Considerations
Critical for source apportionment
Time series of PILS and IMPROVE sulfate and nitrate
Time series of PILS data
APPORTIONMENT QUESTIONS
Many issues here but will focus on smoke
Increasing Information Needs
Contribution of Fires to Particulate Carbon
Wildfire
Prescribed Fire Residential Wood Burning
Agricultural Fire
Contributions from Biomass Burning
Biomass burning can have significant primary and secondary particulate carbon contributions
Smoke Impacting Yosemite NP Summer 2002
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
July 14-20 July 21-27 July 28-Aug 3 Aug 4-10 Aug 10-16 Aug 17-23 Aug 24-30 Aug 31-Sep 5
OC
Sou
rce
Con
trib
utio
n (µ
g/m
³)
SOA/Other
Vehicle Emissions
Meat Cooking
Biomass Combustion
SOA from smoke and other sources
Primary Smoke
Aging Rapidly Creates Lots of SOA
5
4
3
2
1OA
En
han
cem
ent
Rat
io
6543210
Elapsed since lights on (hours)
50
40
30
20
10
0
CO
A (
µg m
-3)
120
80
40
0
Wal
l Co
rrec
ted
CO
A (
µg m
-3)
Lights offMeasured OA
Modeled MassPOA Mass
POA MassModeled Mass
SOA via BC Scaling
frac_orig_60_0 frac_orig_73_0 frac_orig_167_0 smps_VFR_mass_med Ocenh_590fit frac_orig_137_0 frac_orig_fl_av_relfilt
5
4
3
2
1OA
En
han
cem
ent
Rat
io
6543210
Elapsed since lights on (hours)
50
40
30
20
10
0
CO
A (
µg m
-3)
120
80
40
0
Wal
l Co
rrec
ted
CO
A (
µg m
-3)
Lights offMeasured OA
Modeled MassPOA Mass
POA MassModeled Mass
SOA via BC Scaling
frac_orig_60_0 frac_orig_73_0 frac_orig_167_0 smps_VFR_mass_med Ocenh_590fit frac_orig_137_0 frac_orig_fl_av_relfilt
Wal
l-lo
ssC
orr
ecte
d (g
/m3 )
Jeameen Baek et al., - Georgia Institute of Technology
Hybrid Source Apportionment Model
Meteorology
Air Quality
Source-compositions (F)
Source-oriented Model (3D Air-quality Model)(CMAQ, CAMx)
Receptor (monitor)
Receptor Model
(CMB, PMF)
Source Impacts
Chemistry
Receptor model C=f(F,S)
Smoke Management Needs for Air Quality Regulations
• Develop an unambiguous routine and cost effective methodology for apportioning primary and secondary carbonaceous compounds in PM2.5 RETROSPECTIVELY to prescribed, wildfire, agricultural fire, and residential wood burning activities– Daily contributions needed for Haze Rule to properly estimate natural
contribution and contribution to worst 20% haze days – Annual and daily contributions needed for PM2.5 and PM10 NAAQS– Long term data needed to assess successes of smoke management
policies
• Similar needs for ozone and reactive nitrogen deposition issues
Smoke Apportion: Receptor Modeling
• PMF type models with IMPROVE data retrieves a smoke/SOA factor – dominate contributor to contemporary carbon in rural areas
IMPROVEData
Receptor Model
Primary + Secondary Smoke + Vegetation SOC
Mobile Source
Source FactorsSource Profiles
Other Sources
Smoke Apportion: Receptor Modeling
• Addition of primary smoke marker species allows the separation of primary smoke from SOC
IMPROVEData
PMF/other
Mobile Source
Source FactorsSource Profiles
Other Sources
Primary Smoke Marker Species
Primary Smoke
Secondary Smoke + Vegetation SOC
Receptor Model
Smoke Apportion: Hybrid Receptor Modeling
• Addition and incorporation of prior source attribution results in a hybrid receptor model can separate both primary and secondary smoke from other sources
IMPROVEData
Hybrid PMF
Mobile Source
Source FactorsSource Profiles
Other Sources
Primary Smoke Marker Species
Primary & Secondary Smoke
Vegetation SOC
Source OrientedTransport Model (All fires)
Receptor Model
Smoke Apportion: Hybrid Receptor Modeling
• By tagging the prior source attributions by the fire type and the fire location, the contributions of fire can be apportioned to specific fire types and locations
IMPROVEData
Hybrid Receptor Model
Mobile Source
Source FactorsSource Profiles
Other Sources
Primary Smoke Marker Species
Primary & Sec Smoke
Other SOC sources
Source OrientedTransport Model + Fire Types
Secondary Smoke Marker Species
Agricultural Fire
Prescribed Fire
Wild Fire
Fraction Biogenic - Summer 2004-05
The summer (June-August) IMPROVE carbon data were partitioned into fossil and biogenic carbon using the derived fossil and biogenic EC/TC ratios
Fraction Biogenic - Winter 2004-06
The summer (December - February) IMPROVE carbon data were partitioned into fossil and biogenic carbon using the derived fossil and biogenic EC/TC ratios
Contribution of Secondary Organic Carbon during the Summer
• Assumes all winter organic carbon is primary– Underestimates the summer secondary particulate carbon
• Assumes that a similar mix of sources contribute to the particulate carbon in the summer and winter. – Impact on estimate is unknown
Secondary TC
Secondary
OC
Biogenic 36% (6.4) 41% (7.3)
Fossil 23% (10) 36% (15)
0
0.1
0.2
0.3
0.4
0.5
SOC / TC SOC / OC
Fra
ctio
n S
eco
nd
ary
OC
Fossil Biogenic
Sources of CarbonPrimary Secondary
Biogenic
• Smoke - Wildfire, Agriculture and residential wood & open burning
• Pollen• Soil• Cooking
• Smoke - Wildfire, Agriculture and residential wood & open burning• Vegetation
Fossil
• Combustion– Mobile (Automobile, Diesel)
– Off Road Mobil
– Oil/gas
– Coal (power generation, industry)
• Combustion–Mobile (Automobile, Diesel)
– Off Road Mobil
– Oil/gas
– Coal (power generation, industry)
• Evaporative loss of
SMOKE MARKER NEEDS
• Minimally need to measure laevoglucose + other primary smoke markers
• Identify secondary makers
• Modeled markers? (various options)
***** Measure with high degree of accuracy **** Measure with reasonable accuracy*** Measure with low accuracy** Research monitoring * Currently cannot do
Note: measurements should be event based for wet deposition and gases and particles measured at least on a 24 hr schedule.
WET GAS PARTICLE Temporal scale(gas/particle)
SO2/SO4-2 ***** **** ***** Min/hr/day/week
NO2/HNO3-/
NO3-
**** ****(***CASTNET) ****(***CASTNET) Min/hr/day/week
NH3/NH4- *** *** *** Min/hr/day/week
Total ON *** ** ** Integrated sample/event based
ONr *(markers)
*** ***(markers)
Min/hr/day/week (gas/part)Event for markers
ONo *(markers)
*** *** Min/hr/day/week (gas/part)Event for markers
ONb *(markers)
** ** Integrated
WHAT DO WE NEED TO MEASURE ?
Monitoring
• NHx at a number of regionally representative sites
• NOy/NOr at a number regionally representative sites
• Wet ON at a number of regionally representative sites
• Smoke markers at a few sites
Meteorology
Air Quality
Source-compositions (F)
Source-oriented Model (3D Air-quality Model)(CMAQ, CAMx)
Receptor (monitor)
Receptor Model
(CMB, PMF)
Source Impacts
Chemistry
Receptor model C=f(F,S)
Need a data base of ground based measurements, satellite data, and model runs for a integrated
analysis
Next Steps?
• Develop denuders to do NH3, NH4, and HNO3/NO3 split (within 6 months – year)
• Establish/examine need for ON measurements at x number of sites?
• Test filter based measurements for gas/particle phase oxidized ON measurements (1+ years)
• Develop marker technology/methodology for apportioning ON in wet deposition (2+ years ?)
• Do the same for reduced gas/particles (time ?)• Implementation schedule (Funding??)
Species Method
NOy (total oxidized nitrogen) Reduction to NO followed by chemiluminescence
NO3- (particulate nitrate) Denuder/filter sampling followed by
ion chromatography
HNO3 (nitric acid vapor) Filter/denuder and followed by ion chromatography.
NH3 (ammonia) Filter/denuder followed by colorimetry or ion chromatography
NH4+ (ammonium) Denuder/filter followed by
colorimetry or ion chromatography
Tabular summary
WHAT’S MISSING: ORGANIC NITROGEN
Northern Rockies Wildfire Impacts
2005 Agricultural Fires
Southern California Fires
Husar et al., 2007
NOTE
• NOy ≡ NO + NO2 + NO3 + 2xN2O5 + HNO3 + HONO + HO2NO2 + RONO2 (organic nitrates such as PAN and alkyl nitrates) + RONO (organic nitrites) + NO3- (particulate nitrate).
• The ecology community defines “Total Reactive Nitrogen” to include N2O which is not reactive in the sense considered here, thus we define “Total Chemically Reactive Nitrogen” to exclude N2O and N2.
Emissions from Different Fire Types
• Wildfire and wildland fire use, “Natural fires”, are the largest sources of smoke, especially in the western United States.
U.S. Acres Burned
0%
20%
40%
60%
80%
100%
2005 2006WildFire Wildland Fire Use
Prescribed Agriculture*
WRAP Region Fire PM2.5 Emissions Scenarios (tpy)
0
500,000
1,000,000
1,500,000
2,000,000
Prescribed 71,421 195,020
Wildland Use 81,505 659,594
Wildfire 1,489,886 504,654
Agricultural 34,571 34,590
N-F Rangeland 15,454 18,643
2002 Actual2018 Projection A -
FLM