NASA ROSES08: Applying NASA Observations, Models and IT for Air Quality Proposal

Air Quality Cluster > Applying NASA Observations, Models and IT for Air Quality Main Page > Proposal | NASA ROSES Solicitation | Context | Resources | Schedule | Forum | Participating Groups

Proposal Cover Page incl.

Proposals submitted electronically through NSPIRES will use the NSPIRES Proposal Cover Page that is available through the World Wide Web at http://nspires.nasaprs.com/. Access for application to a given NRA is provided through a menu entitled "Solicitations" then accessing the link "Open Solicitations." Once completed by the PI, the Proposal Cover Page must be accessed in the NSPIRES system and submitted electronically by the AOR.

If a hard copy submittal is also required, the submitted cover page should then be printed and signed by the AOR. As directed in the solicitation, the signed copy must be submitted with the original copy of the proposal on or before the proposal due date. In addition, reproductions of the signed Proposal Cover Page are used to preface the required printed copies of the proposal.

NSPIRES automatically assigns a unique proposal number to each proposal only after it has been successfully submitted. NASA uses this NSPIRES number throughout the proposal review and selection process to uniquely identify the proposal and its associated electronic data. If no NSPIRES number appears on the Proposal Cover Page, then it has not been properly submitted through the NSPIRES system.

Proposal Summary 300 words

The objective of this project is to develop a decision-support system (DSS) for the implementation of the new Exceptional Event (EE) Rule, which permits States to flag air quality (AQ) data caused by exceptional air pollution, such as forest fires and dust storms. The Rule requires States to provide evidence for and the quantification of exceptional source contributions. Based on the reported evidence, EPA decides if the EE flag is justified.

Preparing and evaluating the EE evidence is a tedious, costly and technically challenging task for the State and EPA offices. A powerful EE DSS tool will be developed that will allow users to (1) explore and analyze data for specific EEs (2) prepare EE flagging reports (3) evaluate and approve the EE reports. For the States, the powerful EE tools will make the event documentation easy and efficient, while for EPA, the standardized DSS tools will make the decisions more consistent and robust.

The project will achieve its goals primarily by linking, harmonizing and integrating and otherwise ‘connecting the pieces’ contributed by its autonomous core constituent partners represented by the projects GIOVANNI, NAAPS, VIEWS, AIRPACT and DataFed. The a wide range of distributed multi-sensory data (including MODIS, OMI, CALIPSO), suitably processed and packaged for the EE DSS using flexible web service orchestration through DataFed. The EE DSS data browsing, processing, reporting and communication facilities will be combined and presented through a user-friendly portal.

The broader benefits of this are project will include deeper scientific understanding of EEs and innovative application of remote sensing and information technologies to AQ regulatory processes. Building the EE DSS will also contribute to the creation of a persistent core network for supporting AQ applications. The network will also exemplify multi-organization/agency collaboration using the principles and architecture of the Global Observing System of Systems.

Decision-making Activities and Baseline Performance

The proposed DSS is aimed at improving the management of the Nation’s air quality. The quality of ambient air is maintained at healthy levels by the setting and compliance with National Ambient air Quality Standards (NAAQS). Compliance with NAAQS based on measurements using Federal Reference Method (FRM) monitors. In 1996, the NAAQS for PM2.5 was significantly revised by reducing the daily standard from 65 to 35 ug/m3 and recently for ozone from 85 to 75 ppb.

Since the 2006 NAAQS amendments, both PM2.5 and ozone are subject to the new Exceptional Event (EE) Rule which allows the exclusion of data strongly influenced by impacts from "exceptional events," such as smoke from a wildfire or dust from abnormally high winds. States "flag" data for those days that they believe to be impacted by exceptional events. Such flagged days, if concurred with by EPA, may be given special consideration in the compliance calculations. The tightening of the short-term standards and the EE Rule shifts the attention from controlling the yearly average to the reduction and control of short-term, episodic air pollution.

The EE Rule identifies different categories of uncontrollable events: (a) Exceedances Due to Transported Pollution (Transported African, Asian Dust; Smoke from Mexican fires; Smoke & Dust from Mining, Agricultural Emissions) (b) Natural Events (Nat. Disasters.; High Wind Events; Wildland Fires; Stratospheric Ozone; Prescribed Fires) and (c) Chemical Spills and Industrial Accidents; Structural Fires; Terrorist Attack.

Fig. 1.

The decisions related to the Exceptional Event flagging and exclusion are performed at three organizations: States, Regional EPA Offices and Federal EPA, as shown in the schematic Fig. 1. The States need to decide whether a particular sample is to be flagged and prepare a flag justification report. The EPA Regional Offices evaluate the submitted flagged requests and approve/deny the requested flag. The role of the federal EPA is to ensure regional consistency of the flag justification evaluations, resolution of difficult cases and to provide general help interpreting the EE Rule.

Currently, the implementation of the EE Rule is ad hoc and unstructured. The guidelines for preparing the flag justifications are intentionally somewhat vague. As a consequence, the current justifications are highly variable; some States submit very detailed and technical reports while others are brief and descriptive. Also, the EPA Regional Offices currently use ad hoc methods to understand the events, to evaluate the claims, and to make their recommendation. The lack of formal procedures and tools makes evaluation difficult and uneven.

Preparing and evaluating the evidence for flagged data is a technically challenging task both for the State and the Regulatory offices. It requires: Accessing a diverse array of data sources illustrating various aspects of the exceptional event; Integration of the heterogeneous data sources that are frequently incomplete and incompatible; Performing detailed data analysis and to establish "clear causal relationship" between the EE and the increased exceedance and the Rule requires a demonstration that the exceedance would not have occurred but for the presence of the EE. The flagging procedure has to be in accordance with section 40 CFR 50.14 (c)(3)(iii) of the EE rule.

Many State and Regional EPA offices lack the means for executing these challenging tasks. Currently, States are using various ad hoc methods at their disposal, including scanning their monitoring data for anomalous patterns, media reports of fires, dust storms and other EEs. The impact of exceptional sources on the violating monitor site is justified in a qualitative sense. In the past, the EPA evaluation of the State-provided data did not benefit from tools and sources of information which are now available. Fortunately, there are now outstanding opportunities to develop credible and reasonably simple methods for the preparation of flagging documentation through an EE DSS.

The flags and the flag justification reports are prepared by the individual States hence, the primary users of the DSS are the States. However, since the EE flag claims are evaluated by Regional EPA offices, they also represent users of EE DSS. The Federal EPA has a broader range of roles. It develops the NAAQS and the associated Rules, which requires considerable research on the nature of ESs and implementation options. EPA also develops EE reporting templates as guides to the States. The Federal EPA also plays a key role in the design of the EE DSS. In fact, EPA has been supporting and interactively guiding our CAPITA group to explore the design options for EE DSS. The preparation of this NASA ROSE proposal has also benefited greatly from the support, the ideas and the feedback from the evolving EE DSS.

The preparation of the qualitative reports is currently time consuming. One rough estimate provided by an officer of the Federal EPA is that currently it takes about a week of State analyst's time to prepare an EE report. Currently there are hundreds of flagged data samples, of which justification requires several person-years of effort. As the implementation of the EE Rule proceeds and the States get more familiar with data exclusion procedures, it is anticipated that the number of flagged samples will increase by at least an order of magnitude to thousands of flagged samples per year. The proposed EE DSS is anticipated to reduce the report preparation time from about one week to less than four hours per flag. This factor of 10 time-savings can then be used more prudently on analyzing and understanding the State's air quality pattern or exploring mitigation options. Furthermore, the quality of EE flag evaluation decisions will be more objective and uniform.

Earth Science Research Results (1page)

The EE Rule offers outstanding opportunity to infuse NASA data products and information technologies deep into EPA's operational activities on managing air quality. The global-scale, high spatial resolution satellites remote sensing data are particularly suitable for detecting and quantifying natural and manmade air pollution events. (dust smoke, haze). The intense aerosol and gaseous pollutant signal during these events have made satellites indispensable in detecting and following the evolution of such events. Additional AQ benefits of satellites is on evaluating and improving emission inventories for NOx, biogenic VOC and particulates.

Unfortunately, until recently, the role of satellites in EPAs air quality regulatory process was very modes. In fact, the EE Rule is the only air quality regulation that we are aware of, where the use of satellites is explicitly encouraged, see Federal Register (Ref): Information demonstrating the occurrence of the event.. ... satellite-derived pixels indicating the presence of fires; satellite images of the dispersing smoke; Identification of the spatial pattern of the affected area (the size, shape, and area of geographic coverage). This could include, for instance, the use of satellite or surface measurement data; The simplest demonstrations could consist of newspaper accounts or satellite images to demonstrate that an event occurred…

The recent past and anticipated future use of satellite data is succinctly stated in his letter of support by N. Frank, the lead EPA scientist responsible for the development of the EE Rule: "... the fusion of satellite-derived measurements from its multiple sensors, combined with ambient air pollution measurements, meteorological data, and modeled estimates have recently been shown to be very valuable to separate the complex sources of air pollution into anthropogenic and natural components and for understanding when events are allowed to be judged exceptional.".

From the point of view of the proposed work, it is also important that NASA-supported IT, particularly the Service Orientated Architecture and Service Orchestration is also directly applicable to the development and implementation of a broadly usable decision support system. These technologies will markedly improve the quality of EE flagging process and also help the implementation of tools for EE Anomaly Detection, Surface-Satellite Data Fusion and Event Climatology Analysis.

The incorporation of NASA data products into the EE DSS is primarily through the rich capabilities and keen interest of the Co-I partners of this project. The GIOVANNI data portal is a key is a key access portal to the most widely used satellite products in AQ analysis, including MODIS and MISR AOT, OMI and more recently CALIPSO. GIVANNI also provides an array of useful data processing and fusion services. In the NAAPS global aerosol model a number of NASA datasets are assimilated used for validation. MODIS-derived fire location is derived hourly in real-time and converted into model-relevant emissions. The MODIS aerosol optical depth product is operationally assimilated into NAAPS. The MODIS Dust Enhancement Product is used to identify dust sources globally for NAAPS. The AIRPACT modeling system actively pursues the verification of CMAQ model with OMI columnar data for urban-industrial as well as for major fire emissions.

Transition Approach (1p)

The transition of this project will be a smooth and natural completion of our research group's participation in the EE Rule evolution. Since 1998, the PI and his co-workers have facilitated or participated in dozens of air pollution event analyzes, most notably the "Asian Dust Events of April 1998", which documented exceptional impacts of Asian dust on Western North America. The analyzes of Central American Smoke of May 1998 caused record PM2.5 concentrations over much of Eastern U.S. and prompted EPA to issue the first set of guidelines (ref) (precursor to the EE Rule) on the treatment of EEs in compliance calculations. Recent EE analysis examples include the impact of Georgia Smoke on sites in the Eastern U.S. 2007, (Nitrate Event, Quebec Smoke, CA Smoke?). Following the EPA's request the CAPITA group has actively during 2005-2006? participated in the development of EE analysis methods and contributed through exploratory illustrations of the candidate EE analysis methods. These were included in the Docket as supporting documentation for the EE Rule.

After the formal publication of the EE Rule in the Federal Register (ref, ) the CAPITA group was again asked to provide further illustrations of the methods that satisfy the EE Rule (ref wiki). The experience from both projects has clearly demonstrated that satisfying the regulatory requirements of the EE Rule can be supported by a suitable DSS information system (NASA ReASON). The need for such a support system has been strongly voiced by the supervising EPA officer and seconded by regional and State analysts who have seen and used those tools. (ref - Region 4 Georgia ??)

The transition of this AQ decision support project into a persistent operation is best expressed and illustrated in his letter of support by R. Poirot, CT Air Quality Planner, and also member of EPA’s Clean Air Science Advisory Board, Co-Chair of RPOs Monitoring and Analysis Committee etc.. “It is especially gratifying to see that NEDS will build directly on the existing DataFed infrastructure and utilize several related applications including the VIEWS, FASTNET and CATT tools which were specifically requested by and developed for the multi-state Regional Planning Organizations (RPOs). These “user designed” RPO data acquisition and analysis tools continue to attract and support a dynamic, collaborative network of empowered data analysts. By adding better connections to various NASA data products like GIOVANI (and associated NASA science expertise), and adding other perspectives such as quantitative estimates of intercontinental smoke, dust and sulfate impacts from the NAAPS global aerosol forecast model and regional impacts from the AIRPACT forecast model, the NEDS project will substantially enhance the power and use of these existing analysis tools and provide invaluable assistance to state and EPA Air managers for implementing the complex new EE Rule. I look foreword to collaboration on this project in the near future."

Specific activities in the transition phase will include workshops and instruction sessions that will include the State, Regional and Federal AQ managers as users of the EE DSS system. The support will also include extensive web-based instructions provided through the EE DSS community workspace. In the past, user workshops will be held on the use of FASTNet, CATT and other tools. In this project, special effort will be placed on harnessing the contributions of the partners.

In the future... Event occurs, in real time analysts are notified AQ monitoring consoles as well as by news, blog and media reports; Analysts choose tools and data and do further analysis; Evidence (higher quality integrated data) produced by the analysts is published with description in near-realtime as event characterization to be used by other applications including EE DSS; When the EE DSS is triggered and the state analyst chooses to flag the station for a given date then further analysis needs to be done using both EE-specific and generic. Analysts chooses tools to provide specific evidence; When the EE DSS is triggered and the state analyst chooses to flag the station for a given date then further analysis needs to be done using both EE-specific and generic. Analysts chooses tools to provide specific evidence

Performance Measures (1p)

• Usage -
  • # of States
  • Cost Savings
  • Citation Index

Because this is an application REASoN proposal, tracking the extent of data and service use is vital to ensure project work is pertinent. The usage metrics is perhaps the most important of the three classes and improvements in usage metrics inherently includes improvements in interoperability and technology metrics. A range of metrics will be used: Data Access Counts – The wrapper-based federated data access system will allow the counting of data accesses in fine detail and will provide valuable feedback on useful data classes. User Group Membership – As described in the attached letters of support, we have initial interest from the EPA and RPOs in the information system. A key metric in evaluating the performance of this system and progress toward achieving project objectives will be the number of members registered in the community website as well as sub-metrics that measure their level of participation in community activities, such as number of website visits, dialogue participation, submission of bugs and wishes, registration of datasets and tools. The user group membership will also be assessed by the numbers of attendees to the planned workshops during project years 3, 4 and 5.

Cost savings in data use- This metric can be approached using two methods. 1) For those end users who had not used remote sensing data prior to the information services due to prohibitive cost, we can quantify the difference in cost between the estimated prohibitive level and the costs associated with using the developed information services. 2) For those users who have been using NASA data on a consistent basis both before and after the system, we can quantify the cost savings by the difference between costs incurred by the end user both before and after the system was implemented.

These same user groups can be surveyed to determine if there was a change in data quality or in the quality of their own products and decision support system. The groups can also be surveyed to determine new capability gains by end users and user satisfaction. Surveys will likely be conducted during the planned workshops.

Citation Indices –Citations to published material and cross-linking to material on our websites offer valuable metrics in assessing the added value generated by the project. We will also measure the publications that relied on the federated system. For example, during the 1998 Asian dust storm a virtual community that gathered on CAPITA’s dust website (http://capita.wustl.edu/Asia-FarEast/) resulted in 12 papers by international authors in a special issue in the Journal of Geophysical Research dedicated to dust and the event.

The preparation of the qualitative reports is currently time consuming. One rough estimate provided by an officer of the Federal EPA is that currently it takes about a week of State analyst's time to prepare an EE report. Currently there are hundreds of flagged data samples, of which justification requires several person-years of effort. As the implementation of the EE Rule proceeds and the States get more familiar with data exclusion procedures, it is anticipated that the number of flagged samples will increase by at least an order of magnitude to thousands of flagged samples per year. The proposed EE DSS is anticipated to reduce the report preparation time from about one week to less than four hours per flag. This factor of 10 time-savings can then be used more prudently on analyzing and understanding the State's air quality pattern or exploring mitigation options. Furthermore, the quality of EE flag evaluation decisions will be more objective and uniform.

This section must articulate the measures (both quantitative and qualitative) the team will use to determine the outcomes, results, and value of the project. The measures should establish the potential improved performance achieved through integration of the Earth science research results. The measures should include those that the decision makers employ, as well as those used to establish the baseline performance.

How many flags have been performed so far 
Each flag costing week before and then new DSS tools would be hours

Tools allow States to evaluate their own performance better
States can learn from each other using common tools and comparing notes

• The performance of the EE DSS is measurable by the current number of requests for waivers and by the time required for the preparation and evaluation of the requests.
• The improved quality of the decision-making system can be measured by the ratio of the requested to the approved EE flags
• The broader applications can be measured by the number of other DSSs that utilize some of the components of the EE DSS.
• Google Analytics

Anticipated Results (1p)

Preparing and evaluating the EE evidence is a tedious, costly and technically challenging task for the State and EPA offices. The hypothesis is that a powerful EE DSS tool will allow users to (1) explore and analyze data for specific EEs (2) prepare EE flagging reports (3) evaluate and approve the EE reports. For the States, the powerful EE tools will make the event documentation easy and efficient, while for EPA, the standardized DSS tools will make the decisions more consistent and robust.

The anticipated results of this project from the perspective of a State Air Quality Analyst is well-stated in his attached letter of support by R. Poirot of the Vermont Department of Environmental Conservation. "The NEDS project will provide direct and much needed support to State and EPA Air Quality Management Agencies as they work to better understand and implement EPA’s new Exceptional Event Rule (recently rendered much more critical by the newer and tighter daily standards for PM2.5 and ozone). In addition to this DSS support, I believe there will also be multiple “ancillary benefits” that result from NEDS, since in the course of identifying and documenting the relatively few events which are ultimately designated “exceptional” by EPA’s current rule, we - the networked teams of State, EPA, NASA and academic air quality analysts - will inevitably come to a better understanding of the nature and causes of many air pollution events of varying causes, spatial and temporal extents, and degrees of severity. This will aid the development of improved emission inventories, improve estimates of air quality model boundary conditions, lead to better model performance evaluation criteria for dispersion and receptor models, and provide valuable insights to air quality forecasters and health effects researchers. State Agencies will be better able to focus State Implementation Plan (SIP) control strategies for PM, ozone and regional haze on sources which are “jurisdictionally controllable” at the State level; utilize regional, national or international forums for synoptic-scale transport events; and predict, recognize and track uncontrollable events which result from natural sources."

The broader benefits of this are project will include deeper scientific understanding of EEs and innovative application of remote sensing and information technologies to AQ regulatory processes. Building the EE DSS will also contribute to the creation of a persistent core network for supporting AQ applications. The network will also exemplify multi-organization/agency collaboration using the principles and architecture of the Global Observing System of Systems.

• The DSS tools for EEs will provide a formal venue for the incorporation of NASA Earth Science observations and models into AQ regulatory processes
• For the States, the powerful EE tools will make the Event documentation easy and efficient.
• For EPA, the standardized DSS tools will make the decisions more consistent and robust.
• The data flow infrastructure for the EE DSS will also have broader benefits for the implementation of other SOA-based DSSs.

• Extract stuff from four letters of support - Neil, Rich, George, Region.

This section must describe the expected results from the project. This section must state the team’s hypothesis for the expected quantitative improvement(s) over the "baseline" performance. This section should estimate the expected improvement(s) in decision-making enabled from the enhanced (new) decision-making activity and the associated socioeconomic benefits from the improved decision-making. This section should be as specific and quantitative as possible. This section should identify how the expected results will contribute to the specific goals and objectives of the related application area(s).

Project Management (1p)

• EE DSS Specification and Implementation overseen by an advisory group; Performed by end users, providers and mediators
• Core network of end users, providers and mediators
• Ad hoc community participation through open project management

The project will achieve its goals primarily by linking, harmonizing and integrating and otherwise ‘connecting the pieces’ contributed by its autonomous core constituent partners represented by the projects GIOVANNI, NAAPS, VIEWS, AIRPACT and DataFed. The a wide range of distributed multi-sensory data (including MODIS, OMI, CALIPSO), suitably processed and packaged for the EE DSS using flexible web service orchestration through DataFed. The EE DSS data browsing, processing, reporting and communication facilities will be combined and presented through a user-friendly portal.

Yr 1

• Detailed specification and design of EE DSS
• Establish Interoperable Core EE Network
• Develop Core DSS with Fed. EPA

Yr 2

• Seek feedback from States/Regions/Fed
• Complete EE DSS including User Interface
• Tutorials with End Users

Yr 3

• Tech Support to operational EE DSS
• Document, measure performance, evaluate EE DSS
• Applications, testing with other DSS

PI

Kari

Co-I's

Collaborators

Statement of work

This section should articulate the management approach and structure; plan of work; partnership arrangements; and the expected contribution, roles, and responsibilities of the team members. Project schedule and milestones must be included. Note: Meetings (number of, frequency of, etc.) do not qualify as project management milestones.

Statements of Commitment – Co-Is (as needed)

Every Co-Investigator and Collaborator (ref. definitions in Section 1.4.2) identified as a participant on the proposal’s cover page and/or in the proposal’s Scientific/Technical/Management Section must submit a brief, signed statement of commitment that acknowledges his/her intended participation in the proposed effort. In the case of more than one Co-I or Collaborator, a single statement signed by all participants may be submitted. In any case, each statement must be addressed to the PI, may be a facsimile of an original statement or the copy of an email (the latter must have sufficient information to unambiguously identify the sender), and is required even if the Co-I or Collaborator is from the proposing organization. An example of such a statement follows:

"I (we) acknowledge that I (we) am (are) identified by name as Co-Investigator(s) [and/or Collaborator(s)] to the investigation, entitled <name of proposal>, that is submitted by <name of Principal Investigator> to the NASA Research Announcement <alpha-numeric identifier>, and that I (we) intend to carry out all responsibilities identified for me (us) in this proposal. I (we) understand that the extent and justification of my (our) participation as stated in this proposal will be considered during peer review in determining in part the merits of this proposal. I (we) have read the entire proposal, including the management plan and budget, and I (we) agree that the proposal correctly describes my (our) commitment to the proposed investigation."

Letters from End-User Organizations (4 1page letters)

In addition to the brief statements from Co-Is required per Section 2.3.10 of the NASA Guidebook for Proposers, this section may include up to 4, one-page letters from the end-user organizations that will benefit from the proposed project. The letters may include input from the community and beneficiaries served by the end user organizations. All statements or letters must be addressed to the PI and included in the proposal.

Budget Justification: Narrative and Details (1page)

Facilities and Equipment

The final element of the Budget Narrative is a description of any required facilities and equipment. This section should describe any existing facilities and equipment that are required for the proposed investigation. It must explain the need for items costing more than $5,000, describe the basis for estimated cost, and provide supporting documentation to support the estimate (i.e., competitive quotes obtained, justification for sole source purchase, proposed cost based on previous purchases for same or similar item(s), cost data obtained from internet research, etc.).

General-purpose equipment (i.e., personal computers and/or commercial software) is not allowable as a direct cost unless specifically approved by the NASA Award Officer.

Current/Pending Support

Information must be provided for all ongoing and pending projects and proposals that involve the proposing PI. This information is also required for any Co-Is who are proposed to perform a significant share (>10 percent) of the proposed work.

All current project support from whatever source (e.g., Federal, State, local or foreign government agencies, public or private foundations, industrial or other commercial organizations) must be listed. This information must also be provided for all pending proposals already submitted or submitted concurrently to other possible sponsors. Do not include the current proposal on the list of pending proposals unless it has been submitted to another possible sponsor.

All projects or activities requiring a portion of the investigators’ time during the period of the proposed effort must be included, even if they receive no salary support from the project(s). The total award amount for the entire award period covered (including indirect costs) must be shown, as well as the number of person-months per year to be devoted to the project for each year, regardless of source of support.

Specifically, for the PI and any Co-Is who are proposed to perform a significant share (>10%) of the proposed work, provide the following information:

• Title of award or project title;
• Name of PI on award;
• Program name (if appropriate) and sponsoring agency or organization, including a point of contact with his/her telephone number and email address;
• Performance period;
• Total budget; and
• Commitment by PI or Co-I in terms of person-months per year for each year.

For pending research proposals involving substantially the same kind of research as that being proposed to NASA in this proposal, the proposing PI must notify the NASA Program Officer identified for the NRA immediately of any successful proposals that are awarded any time after the proposal due date and until the time that NASA’s selections are announced.

References and citations

All references and citations given in the Scientific/Technical/Management Section must be provided using easily understood, standard abbreviations for journals and complete names for books. It is highly preferred but not required that these references include the full title of the cited paper or report.