1 of 31

Agricultural Soil Emissions of Reactive Nitrogen and their impacts on Air Quality, Health, and Climate

Daniel Cohan and Lina Luo

Rice University

Presentation to U.S. EPA

August 25, 2022

2 of 31

About us

Daniel Cohan

  • Associate Professor of Civil and Environmental Engineering at Rice University
  • EPA Board of Scientific Counselors, but speaking in personal capacity
  • Author of Confronting Climate Gridlock: How Diplomacy, Technology and Policy Can Unlock a Clean Energy Future (Yale University Press, 2022)

Lina Luo

  • Ph.D. student in Civil and Environmental Engineering
  • Lead author of study

2

3 of 31

3

Reactive Nitrogen Emissions from Soils

  • N2O 🡪 climate
  • NOx & NH3 🡪 O3 & PM
  • Leading contributor to PM health effects in US (Tschofen et al., PNAS 2019)
  • Spatially variable

4 of 31

Importance of Soil NOx

  • Large and uncertain source
    • 5-20 Tg N/yr emissions globally
    • ~20% of emissions in California?
  • Temporal variability enhances contribution to summer peak O3 and sensitivity to climate
    • Peaks after fertilizer application
    • Increases nonlinearly with T
    • Pulses when rain follows dry period
  • Declining fossil emissions make ozone more sensitive to NOx

4

BDSNP

Wang

Global Emissions

California Emissions

Sensitivity to Soil Temperature

5 of 31

Importance of Soil Ammonia

  • Leading source of NH3
  • Potent contributor to PM2.5
    • Contributes to both ammonium nitrate and ammonium sulfates
  • PM2.5 causes more deaths than all other air pollutants combined

5

Economic Damage of U.S. Emissions by Sector

Soil Ammonia Emissions

6 of 31

Importance of Soil N2O

  • Leading agricultural source of greenhouse gas emissions
  • Atmospheric lifetime: 109 years
  • GWP100: 273 (IPCC 2021)

6

Sources of N2O

Agricultural Emissions of Greenhouse Gases

Global Mean N2O

7 of 31

Agroecosystem Modeling of Nitrogen Emissions

7

  • Meteorology
  • Soil
  • Topography
  • Crop type
  • Farming practices

Agroecosystem Model

Volatilization

Nitrification

Denitrification

Other processes

 

 

 

8 of 31

8

Prior soil NO studies by Cohan group

(Rasool et al., 2016 and 2019)

Latest study: Soil NOx, NH3, and N2O

and their impacts

(Luo et al., 2022)

9 of 31

Integrated Assessment with FEST-C* and APEEP

9

Enhanced Fertilizer Emissions Scenarios Tool for CMAQ (FEST-C*)

 Air Pollution Emission Experiments and Policy Analysis (APEEP) 

NH3

NOx

N2O

Social Cost

of N2O

Climate Impacts

FEST-C*: Agroecosystem model adapted by Rice from EPIC and DayCent

APEEP: Reduced form air quality & health model

Luo et al., Environmental Science & Technology, 2022

Emissions Estimates

Health Impacts via O3 & PM

10 of 31

Comparison of Estimation Approaches for Nr Emissions

10

Source

U.S. EPA

U.S. EPA

U.S. EPA

This study

Species

NOx

NH3

N2O

NOx, NH3, N2O

Model

BEIS YL

CMU Ammonia Model

DayCent

& IPCC Tier1

FEST-C*

Approach

Parametric

model

Emission factor method

Mechanistic model (DayCent) & Emission factor method (IPCC Tier1)

Mechanistic model

Biome

Soil

Agricultural soil

Agricultural soil

Agricultural soil

Variables

Soil properties, meteorological conditions

Fertilizer input

Soil properties, meteorological conditions, farming practices

Soil properties, meteorological conditions, farming practices

*Note: U.S. EPA uses CMU Ammonia Model for NH3 estimates in 2011 and bidirectional exchange model in 2017

11 of 31

11

Enhanced Features of FEST-C*

Enhanced Fertilizer Use Rates

(based on USGS data)

Original FEST-C

(optimized for crop N demand)

USGS fertilizer

sales data

FEST-C*: Scaled up by USGS data where applicable

Enhanced N scheme

(based on DayCent model)

12 of 31

Comparison of nitrogen budgets in FEST-C and FEST-C*

12

Our modifications to FEST-C increased N input from fertilization by 0.13% and N loss by 14%, leading to a 7% decline in the soil N pool and a less than 1% decline in harvested N

13 of 31

Simulated 3 Years with Different Palmer Drought Severity Index

13

Drought

2011

2012

2017

May

June

July

Moist

14 of 31

Comparison of Fertilizer Use in FEST-C & FEST-C*

14

2011

2012

2017

FEST-C

(based on optimization)

FEST-C*

(adjusted by USGS sales data)

24% increase

26% increase

29% increase

Difference

(to account for overuse)

15 of 31

Comparisons of Soil N Emissions Estimates

15

Estimated agricultural emissions in BEIS YL

  • Our U.S. estimates: ~1 million tons/year each for NOx and N2O; ~1.5 million tons/year for NH3
  • Similar to EPA for NOx and N2O; higher than EPA for NH3
  • Modest interannual variability nationally, but more variable within regions

16 of 31

County-Level Annual Emissions Estimated by FEST-C*

16

NOx

NH3

N2O

2011

2012

2017

Emissions largely follow fertilizer use, but with subtle differences by species and year (e.g., drought)

17 of 31

Ratio of HONO Emissions to NOx (HONO + NO) from Nitrification

17

Maximum HONO/NOx estimates by FEST-C* : 0.35 ~ 0.43 in Oswald et al.

2011

2012

2017

18 of 31

Interannual Variability of Fertilizer N & Soil Nr Emissions

18

N2O

NH3

NOx

Fertilizer N

%change

2012 vs. 2011

%change

2017 vs. 2011

Interannual variability of Nr emissions cannot be fully explained by changes in fertilizer use

19 of 31

Pulsing factors for NOx emissions when follows dry period

19

2011

2012

2017

Intense pulsing of NOx in 2012 across Corn Belt and Northeast regions

20 of 31

Monetized Impacts of Agricultural Soil Emissions by County

20

NOx

NH3

N2O

Marginal damage

($/ton)

Total damage

($)

Luo et al., Environmental Science & Technology

In most counties, NH3 emissions have largest impact

21 of 31

Marginal Health Damages of NOx Emissions in 2011

21

O3

PM

Health impacts via PM dwarf those via O3

22 of 31

Seasonality of Emissions and Impacts

Here, Winter = Q1 (January – March); Spring = Q2 (April – June); Summer = Q3 (July – September); Fall = Q4 (October – December)

Fertilizer use peaks in spring;

Emissions peak in spring & summer

Marginal impacts peak in winter, when PM is most sensitive to emissions

Total impacts by season

23 of 31

Regional Air Quality and Climate Impacts

23

Luo et al., Environmental Science & Technology

  • NH3 (via PM & health) has largest impacts
  • N2O (via climate) comes closer in rural regions
  • NOx (via PM & O3) comes close where PM and O3 are NOx-limited

24 of 31

Forthcoming work: How soil carbon amendments influence emissions

24

ZVC

25 of 31

Potential Impacts of Soil Amendments (good or bad?)

25

Air

Climate

Water

Soil

  • NH3 & NOx emissions
    • Particulate matter
    • Ozone
    • Health
  • Runoff & leaching
    • Groundwater
    • Surface water
    • Eutrophication
  • N2O emissions and carbon uptake
    • Impacts on climate
  • Soil properties
    • Crop yields
    • Fertilizer and water needs

26 of 31

Initial Modeling of Biochar with FEST-C*

  • Simulation year: 2011
  • Domain: U.S. croplands
  • Biochar application
    • Type: hardwood biochar
    • Tillage depth: 15 cm
    • Rate: 5 ton/ha or 20 ton/ha
    • CEC: 185 cmolc/kg
    • Carbon content: 72.9%
    • Bulk density: 0.64 ton/m3

26

  • Cation exchange capacity (CEC): ↑
  • pH: ↑ (less acidic)
  • Carbon pool allocation
    • Put 2% for active/metabolic pool, 60% to slow pool and 38% to passive pool
  • Bulk density

Biochar Algorithm for FEST-C (Lychuk et al. 2014)

Simulation Configurations

How should we change this to model VACS??

27 of 31

27

Impact of Biochar on Soil pH and Cation Exchange Capacity

Base case

Biochar (5 ton/ha)

Difference (Biochar – Base)

Soil pH

Soil CEC

Soil pH and CEC both increase with addition of biochar;

Largest impact on pH in Southeast states

28 of 31

Preliminary Results: Impacts may depend on dose

28

% Change

Increase

Decrease

N2O

NH3

NOx

Impacts of 20 ton/ha biochar

Preliminary Results

Impacts of 5 ton/ha biochar

Preliminary finding: Emissions may increase with low dose biochar but decrease with high dose

Preliminary Results

29 of 31

29

Further Explorations: Modeling Potential Environmental Impacts of Zero-Valent Carbon (ZVC)* Soil Amendment

Represent ZVC and its impacts on soil properties based on laboratory studies

      • Soil pH, CEC, and other properties
      • Allocation of C to active and passive pools

Integrated assessment of impacts

    • NH3, NOx, and N2O emissions
    • Crop yields
    • Water and fertilizer needs
    • Carbon sequestration
    • Compare to biochar

Represent impacts of ZVC on microbial activity

    • Incubation studies

*ZVC: Produced along with hydrogen from methane pyrolysis (Carbon Hub)

30 of 31

Conclusions

  • Soil emissions of reactive N are important and uncertain, and deserve further attention
  • Ammonia is likely the most damaging pollutant emitted from soils, especially upwind of populated regions
  • Soil carbon amendments may have larger impacts via reactive N than via carbon sequestration
  • Further research is needed to explore impacts of farming practices

30

31 of 31

Acknowledgements

  • Funding:

  • Dr. Kristen Foley (EPA): Meteorology data
  • Dr. Stephen Del Grosso (USDA): Soil N2O data
  • Dr. Jesse Bash, (EPA): NH3 estimation
  • Dr. Dongmei Yang (UNC) and Dr. Limei Ran (USDA): FEST-C model
  • Dr. Luca Doro and Dr. R. César Izaurralde (Texas A&M AgriLife Research) and Dr. Curtis Jones (UMD): EPIC model
  • Dr. Quazi Z. Rasool (PNNL): N cycling scheme
  • Dr. Nick Muller and Peter Tschofen (CMU): APEEP model

31