Summary

During the 20th century, the city of Los Angeles diverted surface water flowing into Owens Lake for water supply, transforming the large, closed-basin, saline lake into a small brine pool surrounded by dry playa. Under high winds, the exposed lakebed produced large amounts of airborne dust, resulting in the highest concentrations of airborne particulate matter with an aerodynamic diameter of 10 micrometers or less (PM₁₀) in the United States. Since 2000, the Los Angeles Department of Water and Power (LADWP), at the direction of the Great Basin Unified Air Pollution Control District (District), has been constructing and implementing dust control measures (DCMs) on the dry lakebed, with the objective of meeting the U.S. Environmental Protection Agency (EPA) National Ambient Air Quality Standards (NAAQS) for PM₁₀ and the PM₁₀ standards set by the state of California. LADWP reported that it has spent $2.1 billion on dust control efforts as of May 2019 and that many of the DCMs require large amounts of water, energy, and maintenance to sustain their performance.¹ Shallow flooding is, by far, the most widespread DCM, by surface area, that is applied at Owens Lake. Other DCMs, such as managed vegetation and gravel, are also applied over several areas on the lakebed, and a few small areas ordered for PM₁₀ management are currently uncontrolled. On average since 2007, water use for dust control required 31 percent of LADWP’s fresh water supplies available at Owens Lake,² with a range of 17 to 51 percent.

In 2014, a Stipulated Judgment agreed to by LADWP and the District³ ended litigation concerning dust control requirements and acknowledged the need “for additional effective DCMs that do not rely on water that can be substituted in areas currently under control or applied in areas ordered to be controlled.” The Judgment also acknowledged “the need to

___________________

¹ Prior to LADWP’s launch of dust mitigation projects in 2000, the District conducted research and testing of DCMs in the 1980s and 1990s.

² Available water is assumed to be the sum of LADWP exports in the Los Angeles Aqueduct and Owens Lake freshwater use for dust control. See Chapter 3.

³ Stipulated Judgment in the matter of the City of Los Angeles v. the California Air Resources Board et al. Superior Court of the State of California, County of Sacramento. Case No. 34-2013-80001451-CU-WM-GDS. Approved by the court on December 30, 2014. See https://gbuapcd.org/Docs/District/AirQualityPlans/SIP_Archive/2014_Stipulated_Judgment_20141230.pdf (accessed January 28, 2020).

balance the requirements to control dust emissions and conserve water with the requirements to minimize impacts to cultural and biological resources.” As part of the Judgment, LADWP and the District agreed to contract with the National Academies of Sciences, Engineering, and Medicine to establish the Owens Lake Scientific Advisory Panel (OLSAP, or the panel) to provide ongoing advice on the reduction of PM₁₀ in the Owens Valley. In addition, the Judgment intends the panel to foster communication and collaboration between the LADWP and the District within this context.

The panel’s first task is to evaluate the effectiveness of alternative DCMs for their dust control and water use. The task includes consideration of the associated energy, environmental, and economic impacts and assessing the durability and reliability of such DCMs (see Box S-1).

In interpreting its task, the panel was informed by the definition of environment provided in the California Environmental Quality Act (CEQA), which encompasses impacts on land, air, water, minerals, flora, fauna, ambient noise, and objects of historical or aesthetic significance. The panel discussed key factors within the broad context of that definition with implications for dust management. The panel assessed 15 DCMs (see Box S-2) that represent a range of mitigation approaches that are either being applied at Owens Lake or at various stages of development.

The panel’s evaluation criteria included reported PM₁₀ control efficiency, water use, capital and operating costs, habitat value, protection of cultural resources, durability, reliability, and other factors.

Because quantitative data were not available in many cases, the panel also used semi-quantitative and qualitative approaches to evaluate the DCMs, as necessary.

OVERALL CONCLUSION

DCMs applied at Owens Lake during the past 20 years have significantly reduced PM₁₀ concentrations in the Owens Valley, although further progress in controlling dust is needed to meet air quality standards. The panel evaluated 15 DCMs based on their potential effectiveness in reducing PM₁₀ emissions, water use, and environmental impacts. Based on available data, none of the control measures has been documented to achieve mandated levels of dust control while substantially reducing water use (compared to shallow flooding) and consistently providing quality habitat, although some measures show promise with the need for additional research and testing. Progress toward these multiple goals, including protection of environmentally sensitive areas by reducing land disturbance from DCMs, can be more effectively achieved through a systems approach that considers outcomes over a large spatial scale and interactions among control measures. To inform these decisions, additional research is needed on individual and hybrid DCMs and on the landscape-scale effects of dust control configurations. Evaluation of operational performance of DCMs should be based on airborne PM₁₀ measurements rather than surrogate measures, such as the percentage of a control area that must be covered by vegetation or surface water. Using a systems approach and evaluating DCM performance with PM₁₀ measurements would promote innovative and hybrid strategies for dust control.

PROGRESS IN MANAGING AIRBORNE PM₁₀

Because of the implementation of DCMs on Owens Lake by the District and LADWP, airborne PM₁₀ concentrations at monitoring locations in the Owens Valley Planning Area have decreased significantly. The number of days that near-lake monitors had exceedances of the NAAQS for PM₁₀ decreased from 49 days in 2002 to 8 in 2018 (the last full year for which data were available at the time of this report). The maximum 24-hr average PM₁₀ concentration decreased from 20,750 micrograms per cubic meter (µg/m³) in 2001 to 728 µg/m³ in 2018. The average NAAQS exceedance PM₁₀ concentration has also decreased from more than 1,000 µg/m³ in 2000 to fewer than 241 µg/m³ in 2018. Based on monitoring data from January to June 2019, 4 days had NAAQS exceedances. The maximum exceedance concentration during that period was 451 µg/m³, and the average exceedance concentration was 280 µg/m³.

The reductions in PM₁₀ concentrations are evidence of the general effectiveness of DCMs implemented on Owens Lake. However, additional progress is needed to meet the NAAQS,

which require that daily average PM₁₀ concentrations not exceed 150 µg/m³ more than an average of once per year over a 3-year period. Meeting the air quality standard imposed by the state of California (maximum daily average PM₁₀ concentrations not to exceed 50 µg/m³) would require even greater emission reductions than for meeting the NAAQS. Although further DCMs will be needed for on-lake sources, off-lake sources (mostly located near the lakebed) continue to challenge the ability to attain the PM₁₀ air quality standards within the Owens Valley Planning Area (the southern Owens Valley, where Owens Lake is situated).

NATURAL RESOURCES AND ENVIRONMENTAL CONTEXT FOR DUST CONTROL

Precipitation and water runoff in the Owens Valley are highly variable, which creates significant challenges for dust management and affects available water supplies for export through the Los Angeles Aqueduct. From 1950 through the mid-1980s, the aqueduct consistently provided at least 300,000 acre-ft/year, with even greater production in the 1970s and early 1980s from groundwater pumping and water diversions. Since about 1994, exported flows have averaged 250,000 acre-ft/year, with wide interannual variability. In the past decade, water exports have been significantly reduced (averaging approximately 170,000 acre-ft/year) because of California’s multiyear drought and water needed for dust control at Owens Lake. The 2012-2016 drought required significant changes in both the operation of the aqueduct and dust mitigation strategies. Extreme precipitation has also stressed the ability to manage dust on the lakebed. For example, heavy snow and rapid melt in 2017 threatened to flood portions of the dust control infrastructure and mining operations on the lake.⁴

Climate change is anticipated to adversely impact the Owens Valley water supply and therefore dust control efforts, with longer and more severe droughts and more extreme wet years. Those expected trends will likely reduce the reliability of DCMs that involve the use of large amounts of water. Rising temperatures are also expected to increase evaporation at Owens Lake and thereby increase the demand for water that is used for dust control. At the same time, higher temperatures will reduce average runoff in the Owens Valley, because of increased transpiration by plants and evaporation at higher elevations. Warming temperatures are also likely to increase the dominance of rain over snow in the headwaters of the Owens River watershed, leading to more rapid runoff and increases in spring peak runoff. Because of climate-related changes, the availability of water for dust mitigation will be more variable, more water will be needed during dry periods to mitigate dust and maintain habitat, and more pressure will be put on the system to support downstream water demands.

Dust control efforts at Owens Lake have created a variety of habitats on the lakebed, including now regionally rare habitats such as alkaline meadows and shallow flooded areas.

___________________

⁴ Ore deposits of trona (a salt consisting of sodium carbonate and sodium bicarbonate) are mined at Owens Lake (see Chapter 3).

The diversity of birds supported by those habitats is based on the engineered conditions that vary in water depth, salinity, and surrounding environs. Because highly productive food webs tend to occur in brackish pools, long-term salinity management to maintain these habitats is particularly important. While management of habitats across broad regions will become more challenging under climate change because of habitat loss and effects on breeding and food availability, the habitats provided by Owens Lake will become more critical to local and regional conservation, particularly because saline lake bird habitat elsewhere within the Great Basin is shrinking. In fact, Great Basin shorebird populations have already decreased by 70 percent over the past few decades.

To date, bird populations and habitat have received the most attention, but dust control can be a valuable conservation tool for providing habitat for diverse species. The value of diverse, aquatic and non-aquatic habitats and the relative abundance of those habitats in the Owens Valley are important considerations when setting priorities for lake-wide management decisions.

Local Native American tribes are an integral part of the environment and have a strong sense of ownership and stewardship of land in the Owens Valley. Local tribes have expressed concerns about the potential damage to culturally and historically significant sites from the use of heavy machinery and levelling operations typically used during DCM construction. Tribal concerns also include preservation of the natural landscape, because many topographic features and types of ecosystems are highly valued.

EVALUATION OF DUST CONTROL MEASURES

Based on available data, none of the current Best Available Control Measures (BACMs) or other DCMs has been documented to achieve mandated dust control efficiencies, while substantially reducing water use (compared to shallow flooding) and consistently providing moderate or high habitat values. Of the DCMs reviewed, many involved a high level of land disturbance and infrastructure that could impact cultural resources in environmentally sensitive areas, although a few could be conducted with low land disturbance. Using a variety of data, the panel evaluated 15 DCMs—including approved BACMs and BACM modifications and other DCMs in various levels of testing—for their PM₁₀ control, water use, cost, ability to provide habitat, and other factors, such as the extent of land disturbance and aesthetics. However, the panel did not presume to understand many of the factors that influence the acceptability of a DCM in environmentally sensitive areas. As described in Chapter 4 and summarized in Table 4-1, its findings reveal that no control measure met desired performance in all categories.

Of the DCMs reviewed, precision surface wetting, shrubs, natural porous roughness, and cobbles appear to be promising strategies, individually or in combination, for substantially reducing water use and providing some habitat value (see Chapter 4). For example, field tests suggest that precision surface wetting may be able to provide mandated control efficiency with reduced water use compared to shallow flooding, and refined configurations offer potential

for further reduced water use. The habitat supported by sprinkler irrigation at the test site is similar to that of alkaline meadows, which are regionally rare. However, some amount of shallow flooding is needed to support habitat for shorebirds and Snowy Plover—a species of high conservation value regionally. The use of shrubs as an alternative DCM is promising with lower vegetation cover and water use than currently required by the managed vegetation BACM. Shrubs may not be able to provide mandated control efficiency under entirely rain-fed conditions, but the alternative measure may provide a useful option with some supplemental irrigation or in locations where reduced control efficiencies are allowed. Cobbles and natural porous roughness are promising waterless approaches that have not been tested at Owens Lake and deserve additional attention. Cobbles theoretically offer improved dust control and aesthetics compared to gravel, while allowing native vegetation growth. Natural porous roughness offers improved aesthetics compared to engineered roughness and moderate-value habitat. Initial testing of a solar photovoltaic installation on top of a gravel layer revealed mixed results, but solar panels have dust control potential that could be investigated further, with or without placing the panels on top of a gravel layer. Although it does not provide the same level of habitat value of the other DCMs considered by the panel, the brine BACM provides effective dust control without any freshwater and, when placed in appropriate locations, serves as a sink for salts flushed from other areas on the lakebed. Examples of hybrid DCMs include managed vegetation combined with either artificial roughness elements or precision wetting. The panel did not attempt to judge the acceptability of those DCMs on environmentally sensitive areas.

Recommendation 1: Additional research on individual and hybrid dust control measures (DCMs) should be conducted to develop new approaches that use less water, maximize other environmental benefits, and ensure that DCMs maintain performance over the long term. Specific research topics to inform future decision making at Owens Lake are provided in Chapter 4 and include the following:

• Strategies for long-term salinity management in shallow flooding and managed vegetation DCMs, including the capacity to maintain target salinities over time;
• Minimum percent coverage needed for alternative vegetation species and mixtures of species as DCMs with the potential to reduce irrigation requirements, and how site-specific conditions on the lakebed impact the performance, durability, and management requirements;
• Potential for dynamic precision surface wetting to provide effective control in real-time that reduces water use;
• Approaches for enhancing the formation of salt crusts and their long-term stability under a range of conditions;
• Performance and feasibility of cobbles and natural and artificial porous roughness as DCMs on the lakebed and their potential to provide additional vegetated habitat;

• Potential of hybrid DCMs (such as precision wetting with vegetation) that may lead to further reductions in water use relative to either control measure alone, while increasing habitat value;
• Performance and reliability of current and proposed DCMs under future conditions anticipated from climate change, including longer-term changes in climate and more extreme weather events; and
• PM₁₀ control effectiveness for specific DCMs at various wind speeds.

Rigorous testing and analysis of new alternative DCMs are necessary to develop approaches that demonstrate dust control effectiveness, require less water, and can meet other objectives. This testing should employ improved methods to quantify PM₁₀ emissions, monitor the effectiveness of DCMs, and determine the amount of PM₁₀ emission reductions needed to comply with air quality standards. Improved methods are also needed to assess the effects of DCMs on cultural resources.

Several regulatory constraints limit the capacity to test and transition to new or modified DCMs on Owens Lake. For example, regulations require testing of all new alternative BACMs outside of currently regulated dust control areas and limit the area that can be converted to new BACMs. In addition, current regulations might not allow experimentation with hybrid methods. At the same time, strict regulatory time frames for meeting performance standards can limit the use of managed vegetation, which can be a more sustainable DCM in the long term.

IMPROVING METHODS FOR THE EVALUATION OF DUST CONTROL MEASURES

Quantifying PM₁₀ Emissions

Development of a dust control strategy to attain PM₁₀ air quality standards involves characterizing sources of PM₁₀ emissions and the rate at which each source or area emits PM₁₀ to the atmosphere. PM₁₀ emission rates from individual dust control areas on the Owens Lake bed are estimated using sand flux measurements.⁵ DCMs that are highly effective in reducing horizontal sediment transport are also effective in significantly reducing airborne PM₁₀ concentrations. However, the relationship between PM₁₀ emissions and sand flux is highly variable in space and time, depending on the type and condition of the surface from which PM₁₀ is emitted and meteorological conditions. The variability and uncertainty in the measurements and factors used in that relationship can impart considerable uncertainty to the estimated

___________________

⁵ Sand flux (more generally referred to as horizontal sediment transport) is a measurement of the mass of sand-sized particles per unit time at about 6 inches above a wind-blown surface. Estimating PM₁₀ emissions with sand flux measurements involves the use of a semi-empirical relationship that relies upon the horizontal movement of particles, whose sizes include diameters greater than 10 µm.

PM₁₀ emissions that are fed into air quality models to demonstrate strategies for complying with air quality standards.

Alternative approaches that use direct measurements of PM₁₀ concentrations, made upwind and downwind of dust control areas, can improve the characterization of the level of dust and PM₁₀ control provided by DCMs. Recent advances in instrumentation have enabled the development of low-cost and yet accurate sensors of airborne particulate matter. The networking of these sensors with existing monitors could potentially provide operational managers more accurate and precise PM₁₀ measurements with enhanced spatial and temporal resolution (see Chapter 2).

The accuracy of air quality models would be enhanced by a more direct quantification of PM₁₀ emissions rather than use of horizontal sand flux as a surrogate for PM₁₀ emissions. The importance of accurate estimates of a DCM’s effectiveness in controlling PM₁₀, and associated uncertainties, increases as airborne PM₁₀ concentrations approach the allowable level of the air quality standards. Estimates of reductions in PM₁₀ emissions are associated with a high degree of uncertainty because they have relied primarily on measurements of sand flux.

Recommendation 2: The District and LADWP should develop and apply additional methods to quantify, with uncertainty estimates, PM₁₀ emissions from individual dust control areas, based on direct measurements of airborne PM₁₀ concentrations. All devices should be calibrated and tested for representative operation under the field conditions encountered on and around the Owens Lake bed. Testing should include

• Multiple types of sensors and potential sampling strategies;
• Sites on the lakebed with different soil textures and during different seasons; and
• Proximity to a meteorological site to obtain observations (e.g., humidity and radiation loading) for characterizing local environmental conditions.

In addition, there should be a transition period during which the deployment of a network of PM₁₀ sensors overlaps with the use of the current monitoring network to determine relationships between the historic sand flux measurements and more directly determined PM₁₀ emissions.

Monitoring BACM Effectiveness

When monitoring the effectiveness of deployed BACMs and other DCMs over time, operational managers rely on surrogate metrics (performance criteria) instead of more direct estimates of PM₁₀ emissions. These metrics are derived from data collected for this purpose during the BACM testing and approval phases. Examples include the percentage of a control area that must be covered by vegetation or surface water.

The relationships between the performance criteria and PM₁₀ control efficiencies are, in some cases, based on analysis with highly variable results. Further, the measurements used to determine compliance with performance criteria are, themselves, uncertain. Although estimates of PM₁₀ emissions based on airborne PM₁₀ concentrations can also be uncertain, they are more directly related to the desired outcome of the DCMs than are the surrogate metrics currently used in operational evaluations of DCMs.

Tying operational performance of DCMs directly to airborne PM₁₀ concentrations could enhance the transparency of air quality management planning, provide flexibility to develop innovative and hybrid control methods, and allow for adaptive responses in areas that experience declines in PM₁₀ control efficiency. For example, evaluating performance based on airborne PM₁₀ concentrations rather than on the current criterion of 37 percent vegetative cover could demonstrate that less vegetative cover, with the locations and groupings of particular plants tailored to the site, could achieve the expected emission reductions. This in turn would improve management options depending on site conditions and the type of established vegetation. In addition, use of PM₁₀ emission estimates may enable hybrid approaches, such as developing vegetative cover on shallow flooding water management areas, and adaptive response, such as adding roughness elements to vegetative cover areas experiencing temporary declines.

One disadvantage of relying on control area–specific estimates of PM₁₀ emissions, based on airborne PM₁₀ concentration, is the difficulty in assessing compliance under low to moderate wind conditions. Current surrogate metrics for dust control effectiveness can be evaluated under any wind conditions. To serve as a functionally equivalent replacement for surrogate metrics, PM₁₀ concentrations and emissions must be characterized as a function of wind speed, so that DCM performance can be evaluated at lower wind conditions when air quality standards are not exceeded.

Operational evaluations of BACMs and other DCMs have relied on surrogate metrics to monitor PM₁₀ control efficiency, which introduces a high degree of uncertainty.

Recommendation 3: The District and LADWP should evaluate DCM performance based on PM₁₀ emissions from dust control areas, estimated from measurements of airborne PM₁₀ concentrations under a variety of wind conditions.

Air Quality Modeling

Air quality models play a central role in determining the amount of PM₁₀ emission reductions that will be needed to comply with the NAAQS. The panel recognizes the complexity of the processes that govern PM₁₀ emissions from the Owens Lake area and the subsequent transport and dispersion of those emissions. However, the model used to develop the State Implementation Plan⁶ to demonstrate attainment of the NAAQS for PM₁₀ does

___________________

⁶ A collection of regulations and documents used by a state, territory, or local air district to reduce air pollution in areas that do not meet the NAAQS.

not use state-of-the-art dispersion formulations. Model performance and hence reliability of future projections of PM₁₀ air quality can be improved by paying more attention to the processes that govern emissions and dispersion during high winds, when the highest PM₁₀ concentrations occur.

Recommendation 4: Air quality models to demonstrate attainment of the National Ambient Air Quality Standards (NAAQS) for PM₁₀ should incorporate the current understanding of micrometeorology and dispersion, especially during periods of high winds. Furthermore, the uncertainty associated with modeling those processes should be incorporated into plans to attain the NAAQS.

UNDERSTANDING AND ENHANCING BENEFITS OF A SYSTEMS APPROACH TO OWENS LAKE DUST CONTROL

LADWP and the District seek new approaches to further reduce airborne PM₁₀ concentrations, reduce water use, and meet the requirements of the California State Lands Commission (the main landowner at Owens Lake), the California Department of Fish and Wildlife, and other regulatory agencies, while balancing the concerns of multiple organizations, local tribes, and the general public. Achievement of these goals alone is challenging, and climate change places additional pressures on dust control management at Owens Lake. There is a need to place hybrid or new DCMs in more site-appropriate locations that account for the multifaceted characteristics of the Owens Valley and its environs as an interconnected system.

The complex challenge to meeting the multiple goals related to managing PM₁₀ in Owens Lake can be addressed through a landscape-based, systems approach that is flexible and adaptive. Management of DCMs on the lake has occurred in stages as the DCMs have simultaneously evolved, which has led to constraints and strategies that are not always best suited to the lakebed conditions on which they are applied. The need to replace aging infrastructure and to reduce overall water use provides an opportunity to reevaluate the distribution and landscape-scale design of DCMs on the lake. More careful matching of DCMs to local site conditions (e.g., greater use of managed vegetation on sandy areas that are easily leached by shallow flooding) could achieve long-term dust control with lower water use, lower maintenance costs, and improved salinity management. Landscape-based planning also allows for consideration of control-area size and adjacency issues that could result in reduced water and energy use and improved long-term control. For example, placing brine ponds down-gradient from managed vegetation enables reuse of drainage waters and reduction in pumping costs. Keeping shallow flooding away from managed vegetation (where groundwater salinity is a critical consideration) also reduces pumping requirements. However, the requirement that restricts application of new DCMs to no more than 3 square miles limits the timely transition to more integrated lake-wide dust management practices.

Ultimately, improvements in dust control to reduce PM₁₀ concentrations with lower water use, while protecting environmental resources, will result in tradeoffs that have yet to be fully understood. Such tradeoffs are best evaluated and measured in a systematic way to identify the best selection and application of DCMs and to understand how alteration of one DCM can affect system-level performance. Significant reductions in water use will decrease the areal extent of shallow saline water, which supports a robust food web (e.g., brine flies and shrimp) and provides critical habitat for migrating and breeding shorebirds and waterfowl. Irrigated managed vegetation areas support habitats similar to the alkali meadows that are valuable habitat and were once common in the Owens Valley but are now regionally rare. Some level of water use is necessary to sustain all habitats on the lakebed, although different habitats require different amounts. Current habitat modeling focuses on habitat for multiple bird guilds, without a priority for habitats that are unique along avian flyway corridors or regionally rare. Needed is additional information to support the development of a long-term management plan that aims for integrated, spatially, biologically, and culturally appropriate PM₁₀ emissions control, while accounting for water use, habitat, and preservation of cultural resources (see Chapter 5).

Recommendation 5: To support the development of a landscape-based, systems approach with multiple goals, dust control configurations should be assessed within a lake-wide context, considering long-term management of air quality, surface and groundwater, and salinity; protection of cultural resources; and the regional significance of habitat types and other ecosystems services in the Owens Valley.

FUTURE OLSAP CONSIDERATIONS

As indicated in the 2014 Stipulated Judgment, this report is the first in an expected series of reports by OLSAP. Through continued engagement, OLSAP will provide ongoing assessments and scientific advice on the challenges associated with developing sustainable approaches to reduce dust in the Owens Valley. Through its upcoming activities, the panel could provide valuable advice on implementing the recommendations in this report, especially those regarding the use of PM₁₀ concentration measurements to quantify emissions from control areas and the application of landscape-based, systems approaches to assess dust control configurations at Owens Lake.