New Jersey Clean Air Council Public Hearing The Cumulative Health Impacts of Toxic Air Pollutants on Sensitive Subpopulations and the General Public Status.

New Jersey Clean Air Council Public Hearing

The Cumulative Health Impacts of Toxic Air Pollutants on Sensitive

Subpopulations and the General Public

Status of Cumulative Environmental Impact Methods Development in New Jersey

Steve Anderson

April 13, 2011

2

Outline

• History/Background:– What question is the method designed to

answer?

• Other similar research

• Brief description of current draft method with results

3

History/BackgroundDate Action Key Issues

February 2002

NJDEP Proposed Environmental Equity Rule

-Enhanced public participation in permit process-Screening model used to estimate future impacts

January 2004 NJ Governor Executive Order 96 -Established petition process for communities to self identify-Re-created EJAC

February 2009

NJ Governor Executive Order 131 -Created current EJAC-DEP will review EJAC recommendations for policy and regulatory to consider and incorporate cumulative impacts into its decision-making

March 2009 EJAC Report and Recommendations on Cumulative Impacts

- Recommend DEP develop a screening tool to identify “vulnerable and burdened” communities to help guide various policies and actions

July 2009 NJDEP response to EJACs Recommendations

DEP has developed a preliminary geographic information system-based screening tool

December 2009

NJDEP Posts Current Method on web

After EJAC public meeting and request for comments on method

March 2010 DEP Presentation at EPA Symposium In Washington DC

DEP on panel with EPA and California staff developing similar methods

June 2010 Ironbound Community Corp. EPA CARE Grant

DEP approves ICC to use draft method on a pilot basis as part of grant activities

October 2010 DEP Presentation to NJ Science Advisory Board

Brief Introduction of the method

4

Background

What the Cumulative Impact Method Is

What the Method is NOT

A state-wide screening approach A facility-specific or community level risk analysis

Uses simple indicators of multiple environmental hazards to estimate overall “impact” or “burden”

A scientific risk assessment that quantifies probability that damage to life, health, and/or the environment will occur as a result multiple hazards (e.g. one chance in a hundred)

Compares relative impacts of different geographic areas

Calculates absolute risk to compare to health based standards

“Bias for action” Caution and certainty

5

Methods: Indicators

• Categories of indicators1) Environmental/exposure

- Air exposures, Traffic, contaminated sites,

2) Social/vulnerability- Environmental Justice (race, income)

3) Public health- Asthma, low birth weight

• Current NJDEP method focuses on Environmental Indicators

– We compare environmental to other indicators (correlation)– EPA and other states combine with social, and public health

6

Other similar methods

• EPA Environmental Justice Strategic Enforcement Assessment Tool (EJSEAT)– Only used internally

• Faber (Northeastern University) “Unequal Exposure”

• California EPA report: Cumulative Impacts: Building a Scientific Foundation (Dec 2010)

7

EPA EJ SEAT Strategic Enforcement Assessment Tool

8

Faber (Northeastern University) “Unequal Exposure

9

California EPA

10

Outline of NJDEP ApproachIdentify separate “indicators”

Quantify indicators separately at small geographic scale using GIS

Assess options for combining, weighing or aggregating indicators

Analyze/correlate with other variables

“Scale Up” to larger geographic areas

11

Current NJDEP IndicatorsIndicator Data source Original

Geographic ScaleOriginal Units

NATA cancer risk (1999)

EPA data Census tract Risk per million

NATA diesel (1999) EPA data Census tract Ug/m3

NJDEP Benzene estimate

DEP emission inventory

100 meter grid Ug/m3

Traffic All Congestion Management System

1000 foot buffer Traffic Counts all vehicles

Traffic trucks Congestion Management System

1000 foot buffer Traffic Counts heavy trucks

Density of Major Regulated sites

DEP NJEMS data 100 meter grid Sites per acre

Density of Known Contaminated

DEP SRP data 100 meter grid Sites per acre

Density of Dry Cleaners

DEP GIS data 100 meter grid Sites per acre

Density of Junkyards DEP NJEMS data 100 meter grid Sites per acre

12

Methods: Indicator Quantification• Create 100 meter grid rasters using consistent statewide grid• Calculate statistical z-score for each indicator and grid

– Z score = (value-mean)/standard deviation• Eliminate outliers, z-score >3 are assigned a score of 3

– This impacts less than 0.5% of grids

• Two options used to combine indicators:

– Option 1: Sum all z-scores in each grid • Maximum score of 27 (9 indicators) * (3 max z score)• Quantifies how all indicators impact one area• One or two high indicators can drive results

– Option 2: Count each grid with a z score greater than 1• Maximum score of 9 (9 indicators) * (1 count if z >1 )• Focuses more on higher scores• Highlights areas with multiple high indicators

13

DRAFT Results Option 2: Count of all scores >1

Legend

Grid Impact Count Method

VALUE

0

1

2

3

4

5

6

7

8

9

Counties±0 210,000 420,000105,000 Feet

For Illustration Purposes Only

14

NJ Census Data for Percent Minority

•10 cut points

• Natural breaks

Legend

Counties

Block Groups

Percent Minority

0.00 - 0.07

0.08 - 0.12

0.13 - 0.19

0.20 - 0.28

0.29 - 0.38

0.39 - 0.50

0.51 - 0.63

0.64 - 0.78

0.79 - 0.90

0.91 - 1.00

±0 210,000 420,000105,000 Feet

For Illustration Purposes Only

15

Relationship between Cumulative Impact and Social/Economic Indicators

Figure 1: Relationship Between Cumulative Impact and Percent Minority

00.5

11.5

22.5

33.5

44.5

5

<0.10 0.10 to0.20

0.20 to0.30

0.30 to0.40

0.40 to0.50

0.50 to0.60

0.60 to0.70

0.70 to0.80

0.80 to0.90

>0.90

Percent Minority

Mean

Co

un

t >

1

•Grouped all block groups based on percent minority and poverty

• Calculated average cumulative impact score for combined groups

• Cumulative impact scores increase steadily with increasing percent minority and poverty

Figure 2: Relationship Between Cumulative Impact and Poverty

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

5.00

< 2.5 2.5 to5

5 to7.5

7.5 to10

10 to12.5

12.5to 15

15 to17.5

17.5to 20

20 to25

>25

Percent Poverty

Mean

Co

un

t >

1

16

Work Needed• Updates/improvements to existing indicators

– Example….NATA 2005 results for diesel particulate

2005 Estimated Concentrations in New Jersey

Diesel Particulate Risk

Maximum average census tract concentration is 7.9 ug/m3, or

2379 times the health benchmarkHealth Benchmark = 0.0033 ug/m3

Source Contribution

Point - 0%Nonpoint - 0%On-Road - 47%Nonroad - 53%

Background - 0%Secondary - 0%

1000 - 2400 times benchmark

Under 10 times benchmark

10 - 50 times benchmark

50 - 100 times benchmark

100 - 1000 times benchmark

NJDEP - BTS

Risk Benchmark = 1 in a million

17

2005 Estimated Concentrations in New Jersey

Nonroad Diesel Risk

Maximum average census tractconcentration is 6.71 ug/m3, or

2033 times the health benchmark.Health Benchmark = 0.0033 ug/m3

NJDEP - BTS

Under 10 times benchmark

10 - 50 times benchmark

50 - 100 times benchmark

100 - 1000 times benchmark

1000 - 2100 times benchmark

Risk Benchmark = 1 in a million

18

2005 Estimated Concentrations in New Jersey

Onroad Diesel Risk

Maximum average census tractconcentration is 3.35 ug/m3, or

1015 times the health benchmark.Health Benchmark = 0.0033 ug/m3

NJDEP - BTS

1000 - 2100 times benchmark

Under 10 times benchmark

10 - 50 times benchmark

50 - 100 times benchmark

100 - 1000 times benchmark

Risk Benchmark = 1 in a million

19

Work Needed

• Potential new environmental indicators (Environmental Public Health Tracking)– Drinking Water (Community water systems and Private Well Testing)– Ambient Water Monitoring– SRP soil and groundwater contamination data/RPS?– Ambient Air Monitoring/CMAQ– Radon, Radiation– Facility Release/Emission data– Modeled areas “impacted” where possible to replace where data is

currently release point– Communication/Provide “context” to data

• Vulnerability and Health data– Working with DHSS

• Stakeholder Input