Deposition Monitoring Networks and Data

1
Deposition Monitoring Networks and Data Critical
Loads WorkshopCharlottesville, VA May 24,
2006 Gary LearEPA, Office of Atmospheric
Programs
2

Outline

• Provide an overview of
• The monitoring networks, data and information
  data available for use in critical load
  assessments

• The limitations or considerations in using or
  interpreting results from monitoring networks

3
The National Atmospheric Deposition Program
National Trends Network (NADP/NTN)

• Broad geographic and ecosystem coverage
• Sites are located away from local sources
• Regionally representative locations
• Consistent and comparable measurements
• Comprehensive QA program

4
The National Atmospheric Deposition Program (NADP)
5
Relevant Measurements for Input to Critical Load
Assessments
SO4
N
NH4
Ca
6
NADP/NTN Quality Assurance

• Ensure data are comparable from year-to-year and
  site-to-site
• Extensive Documentation
• External QA Program (USGS)
• External Site Audit Program (EPA)
• Annual completeness criteria ensure
• Sufficient precipitation is analyzed to be
  representative of preciptation at the site
• Collectors are efficiently collecting
  precipitation

7
Trends in Sulfate Deposition
2002-2004
1989-1991
8
NADP/NTN Completeness Criteria
Aerochem collector has poor collection efficiency
in snow Some sites routinely fail completeness
criteria
NADP/NTN sites meeting completeness criteria less
than 50 of all years
CO10 at Gothic, Colorado
9
NADP/NTN Completeness Criteria

• These sites are not represented in annual
  aggregations or maps

Inorganic Nitrogen in Precipitation (2002-2004)
CO10 at Gothic, Colorado
10
NADP/AIRMoN
Atmospheric Integrated Research Monitoring
Network (AIRMoN)
11

CASTNET Overview

• Gas and particle concentrations in air are
  measured weekly on filters and used to estimate
  dry deposition
• An inferential model is used to estimate dry
  deposition rates for ozone, SO2, HNO3, and
  particles
• CASTNET sites are located in rural areas and thus
  provide regionally representative data.
• Data record from 1987
• http//www.epa.gov/castnet

12
CASTNET Measurements
13
Concentrations SO2
2002-2004
1989-1991
14
Concentrations HNO3
2002-2004
1989-1991
15
Estimating Dry Deposition

• Product of concentrations and modeled deposition
  velocity
• Dry deposition velocity is a surface phenomena,
  not a weather phenomena
• Valid only for point locations
• No scientifically-valid method for interpolating
  data

Vd1.7 cm/s
Vd0.3 cm/s
Dry deposition velocity for HNO3 at Beaufort, NC
16
Estimating Dry Deposition
CASTNET is currently using the Multi-layer Model
(MLM)

• Developed from the Big Leaf Model
• 20-layer model in which parameters are modified
  by the redistribution of heat, momentum, and
  pollutants.
• Meteorological input
• Site variables
• Maximum Leaf Area Index (LAI)
• Leaf out
• Vegetative type

17
Multilayer Biochemical Model (MLBC)

• CASTNET will begin testing MLBC in July 2006
• Based on Multilayer Model
• Adds biochemical plant growth module
• Photosynthetically driven stomatal resistance
  (Berry-Farquhar approach)
• Cuticular resistance based on membrane transport
  theory
• Soil resistance and water stress revised from MLM
• Ammonia module
• More information at Wu et al, JGR 1084014, 2003

18
Multilayer Biochemical Model (MLBC)

• Improved performance over MLM in modeling
  observed deposition in time series and diurnal
  variations (Wu et al, JGR 1084014, 2003)

Nashville, TN
Kane Forest, NY
19
Multilayer Biochemical Model (MLBC)

• Improved performance over MLM in modeling
  observed deposition in time series and diurnal
  variations (Wu et al, JGR 1084014, 2003)

Nashville, TN
Kane Forest, NY
20
Estimating Dry Deposition

• Everybody loves complex terrain
• But nobody want to model it
• CASTNET underestimates deposition velocities at
  these sites

CASTNET sites meeting assumptions of MLM and MLBC
21
Estimating Dry Deposition

• Combining NADP/NTN and CASTNET data provides
  estimates of total deposition
• Significant variability within regions

22
Estimating Dry Deposition

• Combining NADP/NTN and CASTNET data provides
  estimates of total deposition
• Significant variability within regions

23
Temporal and Geographic Distributions Are
Important

• Arentsville, PA

Joshua Tree NP, CA
24
Temporal Distribution Is Important
Joshua Tree NP, CA

• Arentsville, PA

SO2
HNO3
25
Temporal Trends in Deposition Velocities
SO2
NC
IL
The common wisdom is that, on average,
deposition velocities remain relatively constant
PA
MD
CA
26
Uncertainties in estimating dry deposition
Case Study Egbert (Ontario) Collocated CASTNET
and CAPMoN Site
27
CASTNet and CAPMoN Air Concentrations
Results CASTNET and CAPMoN SO2, SO4, HNO3 and
NO3- concentrations compare very well
28
CAPMoN and CASTNet Dry Deposition Velocities for
SO2, SO4, HNO3, fine-NO3- and coarse-NO3-
SO2
SO4
HNO3
NO3-
CAPMoN dry deposition fluxes gt CASTNET fluxes
due to differences in the modeled dry
deposition velocities
29
Validating CMAQ with CASTNET Measurements
Flux

• CMAQ 2001, 36km grid
• Pretty good correlation for concentrations, but
  not deposition velocities

SO2
CMAQ
Vd
Conc
CASTNET
30
Validating CMAQ with CASTNET Measurements
Flux

• CMAQ 2001, 36km grid
• Pretty good correlation for concentrations, but
  not deposition velocities

HNO3
Vd
Conc
CMAQ
CASTNET
31
Regional Depiction of Dry and Total Deposition
CMAQ output for total sulfur deposition (2001)
CMAQ output for dry sulfur deposition as
percentage of total sulfur deposition (2001)
32
Regional Depiction of Dry and Total Deposition
CMAQ output for total nitrogen deposition (2001)
CMAQ output for dry nitrogen deposition as
percentage of total nitrogen deposition (2001)
33
Ammonia

• No network data avilable
• Efforts are underway to make passive and active
  ammonia measurements

CMAQ predicts over 30 of total N deposition is
due to ammonia in some areas (2001)
34
Summary

• NADP and CASTNET provide consistent, reliable
  data for input to critical load determinations
• Data are available for inorganic major anions and
  cations in precipitation and dry deposition
• CASTNET dry deposition values are modeled
  approximations and useful for reference but not
  extrapolations or interpolations
• Ammonia deposition is a significant component of
  nitrogen deposition but currently not monitored
  by eaither NADP or CASTNET