Natural Event Documentation

Addendum

Natural Event Documentation

Imperial County, California

June 5, 2007

Imperial County Air Pollution Control District

September 8, 2008

Natural Event Documentation

1.Summary

On June 5, 2007 five Imperial County air monitoring stations recorded PM10 concentrations in excess of the State 24-hour PM10 standard of 50µg/m³ and the Federal 24 Hour PM10 standard of 150µg/m³. These air monitoring stations and its respective 24-hour PM10 concentrations are listed below:

Niland English Road, ARB site number 1200694, 162 g/m³

El Centro9th Street, ARB site number 1200694, 200 g/m³

Westmorland-W 1st Street, ARB site number 1300997, 226 g/m³

Brawley – Main Street, ARB site number 1300693, 281 g/m³

Calexico – Ethel Street, ARBsite number 1300698, 282 g/m³

PM10 exceedances recorded in the ImperialCounty on June 5, 2007 meet the criteria for natural events as defined by federal policies. Several factors such as high winds, extremely high heat and lack of precipitation in theImperial Valley contributed to the exceedances of PM10 concentrations. The extreme wind event affecting this area began in the late morning hours of June 5, 2007, and ended during the late hours of June 6, 2007. The extreme wind conditions took place during a time of year when the soils in the area were very dry (0.27 inches of rain recorded between June 2006 and May 2007) and dust could be entrained into the atmosphere by high winds and wind gusts. Figure 1 presents a map of the region with the location of the monitoring stations and PM10 measurements recorded on June 5, 2007.

The strongest winds and blowing dust affecting the Imperial Valley were observed to the west and northwest regions, indicating a large scale regional event. The blowing dust was transported from the west where deposition occurred in most regions of the ImperialCounty. The ImperialAirport recorded wind speeds as high as 37 mph and wind gusts as high as 47 mph. Excessive winds speeds and wind gusts were also recorded at the Blythe, Yuma and ThermalAirports, all of which are in close proximity to the Imperial Valley.

This report is an addendum to the June 5, 2007 Natural Event Documentation that was submitted to the U.S. EPA by the California Air Resources Board on June 13, 2008. This report further demonstrates that without this exceptional wind event, there would not have been an exceedance of the PM10 National Air Quality Standard on June 5, 2007. The purpose of this addendum is to provide additional analysis to demonstrate that:

The dust event was not reasonably controllable or preventable because the PM10 produced originated from a non-anthropogenic source.

There is a clear causal connection between the exceedance and the wind event.

Figure 1.

Imperial CountyAir Monitoring Stations and PM10

concentrations registered on June 5, 2007

1. Dust Event Not Reasonably Controllable or Preventable

1. AnzaBorregoDesert – Dust Emissive Characteristics.

This section will include additional detailed information on dust emissive characteristics for the AnzaBorregoDesert which is the source of fugitive dust for the June 5, 2007, natural event episode in ImperialCounty.

The AnzaBorregoDesert is a section of the SonoranDesert. The AnzaBorregoDesert covers the eastern portion of San DiegoCounty,the western region of ImperialCounty andthe southern section of RiversideCounty between the eastern slopes of the PeninsularRanges (east of Highway 79) and the Salton Sea. Toward the north it extends eastward across the Santa Rosa Mountains from the town of Anza on Highway 371, while toward the south it extends up to the International border between Mexico and the United States.

In May 2004, ENVIRON prepared for the ICAPCD a report entitled “Development of a Windblown Fugitive Dust Model and Inventory for ImperialCounty” (Windblown Dust Study). The results of the Windblown Dust Study were revised in September 2004 as a result of discussions with CARB and USEPA, as documented in Appendix A of the 2005 BACM report. A copy of this report is attached (Attachment A). The report presents the methodology used in estimating windblown fugitive dust from agricultural and barren lands in theImperialCounty. Chapter 3 presents information regarding land use/land cover and soil characteristics for the desert areas, including the AnzaBorregoDesert. As discussed in this report, the primary source of land use/land cover (LULC) data is based on land use surveys conducted by the California Department of Water Resources (DWR) and augmented with the USGS National Land Cover Database (NLCD).

The DWR Land Use Survey databases are developed by the department’s Division of Planning and Local Assistance in order to aid in the efforts of continually monitoring land use for the purpose of tracking the amount of water consumed and changes in its usage. The data is available at the county level and includes detailed agricultural land use classifications, as well as less detailed urban and native vegetation land uses. The data was compiled from aerial photography alongside extensive field verification and is updated on a rotating basis by the county approximately every seven years. Thus, the data is based on varying years, depending on the county, ranging from the late 1980’s through the late 1990’s. The data is very accurate due to the field survey verifications of the aerial photograph imagery. Because the DWR survey data does not explicitly consider certain land use types (i.e., urban lands, desert lands, forest lands, etc.)it was necessary to augment the DWR land use data with the National Land Cover Data (NCLD) data.

The NLCD was developed as part of a cooperative project between the U.S. Geological Survey and the U.S. Environmental Protection Agency to produce a consistent land cover data layer for the entire conterminous U.S. based on 30-meter Landsat thematic mapper (TM) data. The NLCD was developed from TM data acquired from the Multi-Resolution Land Characterization (MRLC) Consortium, a partnership of federal agencies that produce or use land cover data.

A display of the combined DWR and NLCD landuse database is shown in Figure 2 for the Imperial Valley domain. Figure 2demonstrates that the majority of the AnzaBorregoDesert is covered bysand and shrubland, both of which are characteristics of the desert regions.

Figure 2. Landuse coverage for Imperial Valley based on combined DWR and NLCD information.

To determine the characteristics or type of soil present in ImperialCounty, the Wind Blown Fugitive Dust report used the State Soil Geographic Database (STATSGO). The STATSGO database was developed by the Natural Resources Conservation Service of the US Department of Agricultural (USDA), providing detailed information concerning the taxonomy of the soils, including soil texture class, percentage of sand, silt and clay, and the available water capacity of the soil. Figure 3displays the STATSGO data for the ImperialCountydomain. The data in figure 3 corroborates the information in Figure 2, which indicates that the soil texture at the AnzaBorregoDesert area is composed of sand and sandy loam, making this area unstable and susceptible for soil erosion even at low wind speeds.

Figure 3. STATSGO Soils Database soils texture classification.

1. Emission Sources and Activity Data.

The Imperial County APCD adopted Regulation VIII, Fugitive Dust Rules, onNovember 8, 2005. The adopted Regulation VIII rules fulfill best available control measures for serious PM10 non-attainment areas. The U.S. EPA is currently evaluating the Regulation VIII rules for SIP approval. Regulation VIII requires implementing fugitive dust control measures for source categories such as: Bureau of Land Management (BLM) controlled areas, construction and earthmoving activities, handling and transport of bulk materials, disturbed open areas, paved and unpaved roads, and agricultural operations. Regulation VIII standards are performance based whereby the operators are allowed to determine the control techniques sufficient to limit visible dust emissions to 20 percent opacity and, if applicable to that source, to implement requirements for a stabilized surface.

Agricultural burning was allowed in ImperialCountyfor the dateJune 5, 2007. However, minimum agricultural burning was conducted on June 5, 2007, due to the weather forecast indicated high winds with gusts for this day. The Imperial County APCD authorized agricultural burning for a total of 70 acres in the Calipatria area during this day. A copy of the Imperial County APCD Daily Weather and Agricultural Burning Information report is attached to this report (Attachment B).

The APCD inspectors performed minimum facility inspections on June 5, 2007. APCD staff did not observe any unusual emissions other thanthe event of blowing dust on June 5, 2007. The Imperial county APCD received two dust related complaints which investigated on the following day. The results of the investigation indicated that high winds on the previous day caused dust blowing into the property of the complainants. Based on APCD field staff and industry representatives, the APCD estimates that the anthropogenic emissions were approximately constant before, during and after the event, indicating that the significant increase in PM10 concentrations was caused by the wind entrained dust.

3.Causal Connection between the PM10 NAAQS Exceedance and the Wind Event.

a. Threshold Dust - lofting Wind Speed for Imperial County Desert Areas.

This section will include information on wind speeds necessary for entraining and transporting PM10 particulates.

The Windblown Dust Study presents the methodology to estimate windblown fugitive dust from agricultural and barren lands in ImperialCounty. The average windblown dust emissions from open areas, which include emissions from grassland, dunes, and other barren lands, were estimated at 157.34 tons per day of the 2004 emissions inventory. When included in the emissions inventory, the PM10 emissions from open areas account for just over 55% of the total 2004emissions inventory. The PM10 fugitive emissions were estimated assuming a threshold wind velocity of 15 mph or higher. The Imperial County Wind Blown Fugitive Dust report assumes a wind speed of 15 mph or higher as necessary to erode and entrain fugitive dust. This threshold is more appropriate in unstable areas, such as the AnzaBorregoDesert.

Figures4 and 5represent the maximum hourly wind speed data for theJune PM10 monitoring days for the years 2002 through 2007 for the Brawley and Westmorland stations, respectively. With the exception of the June 5, 2007, which is flagged in AQS as a high wind natural event, maximum hourly wind speeds in June were for the most part between 5 and 15 mph at both stations. With the exception of June 8, 2004, which was also documented as a wind event, and June 5, 2007, the PM10 data shows that PM10 concentration remain below 150 µg/m³ in both stations. The June 8, 2004, high wind event episode was not approved by U.S. EPA due to it was documented after the documentation period allowed by EPA. The data demonstrates that PM10 exceedances on June at Westmorland and Brawley were unusual and the high PM10 readings were caused by high winds.

b.Difference between the wind speed data reported.

This section will discuss the difference between the wind speed data collected at the monitoring sites operated by Imperial County APCD and CARB and the wind data collected at ImperialAirport and El Centro NAS.

The Imperial County APCD and CARB monitoring sites collect hourly averages of wind speed and wind direction. Each hourly average is comprised of 60 data points mathematically averaged every hour.Because of limited monitoring data capacity, the APCD does not currently have the capability to report minute data. On the other hand, the Federal Aviation Administration/National Weather Service (FAA/NWS) stations located in ImperialCounty report wind speed and direction data following the standard METAR observation periods. The wind speed and direction data are averaged following the standard METAR observation period, which is approximately 10 minutes prior to the observation time. In other words, the district’s wind speed data recorded is an hourly average for the complete hour and the FAA/NWS data is a 10 minute average in each hour. In addition, airport monitors have wind gust reporting capabilities not available to county monitoring sites. Because the different averaging periods between the APCD’s monitors and the airport monitors, the wind speed may not always be comparable. However, the APCD utilizes both data sets to further substantiate the wind event.

1. Cause for early PM10 concentrations on June 5, 2007.

This section will explain why hourly PM10 concentrations recorded on June 5, 2007 at Westmorland, Niland, and Brawley began to increase significantly at 10 am when, according to information in Table 6 of the District’s documentation, winds did not exceed 20 mph until 12 noon and 1:00 pm.

A high windblown dust event does not always show an immediate impact on high PM10 concentrations at monitoring stations. A blowing dust event may be comprised of entrainment of dust by high winds in the dust source area, and then deposition of dust in the receptor areas with lower wind speeds. Once dust is suspended upwind of the monitoring stations, it can be carried downwind by winds that are below the dust entrainment wind speed thresholds. The documentation shows that the blowing dust was transported from the west where it was deposited in Westmorland, Niland, and Brawley and other western sectors of ImperialCounty. However, Table 12 shows that in fact, excessive wind speeds and wind gusts were recorded at ImperialAirport at 9:53 am, which indicates high wind activity in the region during the mid-morning. Imperial Airport is in close proximity to the Westmorland and Brawley stations.The Imperial County APCD does not have a monitoring station at the AnzaBorregoDesert area. Therefore, the exact time on which high winds impacted the AnzaBorregoDesert, the source of fugitive dust, is unknown. The high PM10 concentration recorded at Westmorland, Niland, and Brawley during the early hours ofJune 5, 2007can be explained as an effect of the transport of dustoriginating from the AnzaBorregoDesert area.

4.“But For” Analysis

1. Wind Speeds Analysis.

On April 15, 2008, the Imperial County APCD submitted documentation for the June 5, 2007wind event in ImperialCounty. This document contains a detailed analysis of wind speed and wind direction data recorded on June 5, 2007, at the ImperialCounty airport, as well as Blythe, Thermal, and Yuma, AZ, airports. In addition, wind speed and wind direction from the monitoring stations operated by the Imperial County APCD and CIMIS stations are included in this document. Finally, to further complement this report, the Air Resources Board made athoroughanalysis of wind speed and wind direction data for this wind event, demonstrating that violation of the PM10 NAAQS would not occur “but for” fugitive dust generated at the Anza Borrego Desert due to high winds. A copy of these documents is attached for your information and further review.

1. Type of Sources contributing to PM10 NAAQS Violation

In cooperation with CARB, the Imperial County APCD develops a complete PM10 emission inventory every year for all sources in ImperialCounty. An emission inventory is a comprehensive description of the sources of air pollution in a given region, along with a quantitative estimate of their emissions. The PM10 emission inventory compares the type of sources that generate dust and could potentially contribute to PM10 NAAQS violations. The 2004 PM10 emission inventory is summarized in Table 1. The 2004 PM10 emission inventory portrays annual average emissions; however it does not directly reflectemissions during exceptional events. Since the majority of the sources included in the emission inventory are controlled, the PM10 annual average emissions may accurately reflect emissions during windy days, with the exception of natural sources which are chiefly uncontrolled. The 2004 PM10 emission inventory features adjustments made to incorporate revised cattle emissions, revised windblown dust model results, and updates for entrained and windblown unpaved road dust estimates.

Table 1 Revised 2004 Annual Average PM10 Emission Inventory for ImperialCounty(tons per day)
Source Category / 2004 PM10 Emissions
(tons/day)
Fuel Combustion / 0.41
Waste Disposal / 0.00
Cleaning/Surface Coatings / 0.00
Petroleum Prod/Mktg / 0.00
Industrial Processes: / 2.77
Mineral Processes / 2.61
Food/Agriculture / 0.16
Solvent Evaporation / 0.00
Res Fuel Combustion / 0.09
Farming: / 9.89
Tilling / 7.11
Harvest / 0.01
Cattle / 2.77
Paved Road Dust / 4.09
Entrained Unpaved Road Dust: / 61.00
City/County / 26.64
Canal / 31.56
BLM/USFS / 1.39
Farm / 1.41
Windblown Dust: / 200.91
Open Areas: / 157.35
Grassland* / 98.75
Dunes* / 19.85
Other Barren Lands* / 38.74
Urban / 0.01
Unpaved Roads: / 30.96
City/County / 7.82
Canal / 16.76
BLM/USFS / 0.37
Farm / 6.01
Non-Pasture Ag Lands / 10.81
Pasture / 1.79
Construction / 1.91
Fires / 0.00
Waste Burning / 2.19
Cooking / 0.06
On-Road Mobile / 0.38
Other Mobile / 0.69
Total / 284
Total Excluding Uncontrollable Natural Sources / 129

* Natural sources of dust, only 1% of which come from lands disturbed by humans.

As presented in Table 1, one of the largest contributors to the PM10 emission inventory is windblown dust from open areas which accounts for approximately 55% of the total emission inventory. To estimate PM10 emissions from this source category considering the impact of high winds, the Windblown Dust Study estimates monthly PM10 emissions from all open areas, except unpaved roads. Table 2 presents the monthly windblown dust estimates for all areas, except unpaved roads. As presented in table 2, the windblown dust emissions from open areas increase significantly duringJune due to the impact of winds.

Table 2 Monthly Windblown Emissions and Annual Average (tons/day)