UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
JUN 22 I998
OFFICE OF
THE INSPECTOR GENERAL
MEMORANDUM
SUBJECT: Office of Water Data Integration Efforts
Audit Report No. EINWG6-15-0001 -8100177
FROM: Michael Simmons
Deputy Assistant Inspector General for Internal Audits (2421)
TO: Robert Perciasepe
Assistant Administrator for Water (4101)
Mark E. Day. Acting Director
Office of Information Resources Management (3401)
The objectives of our survey were to review the Office of Water's (OW) development of
plans to integrate automated water systems data, and to determine whether Federal, Agency and
OW guidance was provided to ensure use of standard data elements for sharing information. We
also reviewed OW activities to increase the public's access to key water data.
Our survey indicated OW has started to address water data integration, but progress is
slow and a more significant effort is needed. Data management and cross integration of media
has been a long-standing problem for the Agency. In 1994, the Office of Information Resources
Management (OIRM) acknowledged data management as a Federal Managers' Financial
Integrity Act (FMFIA) weakness. The OW has not developed Information Technology (IT)
plans, data standards or standard guidance that will facilitate data sharing or integration of data
across OW systems, except for Order 7500.1 A (dated 10/29/92), Minimum Set of Data Elements
for Ground Water Quality.
In this report, we have incorporated ongoing OW and OIRM initiatives, summarized
Federal and Agency IT policies, identified intergovernmental activities in this area, and
recommended corrective actions which would effectively address the identified weaknesses.
ACTION REQUIRED
In accordance with EPA Order 2750, we have designated the Assistant Administrator for
Water as the action official.for this report. As the action official, you are required to provide us
with a written response within 90 days of the final audit report date. The response should
RecyclBd/Rocyclabla Primed with Veoelabte Oil Based Inks on 100% Recycled Paper (40% Postconsumer)
-------
incorporate actions from the Office of Information Resources Management. For corrective
actions planned, but not completed by the response date, reference to specific milestones dates
will assist in deciding whether to close this report.
This audit report contains findings that describe problems the Office of Inspector General
has identified and corrective actions the OIG recommends. The audit report represents the
opinion of the OIG and the findings do not necessarily represent the final EPA position. Final
determinations on matters in this audit report will be made by EPA managers in accordance with
established EPA audit resolution procedures. Accordingly, the findings described in this audit
report are not binding upon EPA in any enforcement proceeding brought by EPA or the
Department of Justice.
Should you or your staff have any questions regarding this report, please contact
Patricia Hill, Director, ADP Audits and Assistance Staff at (202) 260-3615. Please refer to the
report number on all related correspondence. We have no objection to the further release of this
report to the public.
Attachment
-------
OFFICE OF WATER
DATA INTEGRATION EFFORTS
Report No. E1NWG6-15-0001-8100177
INTRODUCTION
Our survey indicated that the Office of Water (OW) has started to address water data integration,
but progress is slow and a more significant effort is needed. Data management and cross
integration of media has been a long-standing problem for the Agency. In 1994, the Office of
Information Resources Management (OIRM) acknowledged data management as an Agency-
level Federal Managers' Financial Integrity Act (FMFIA) weakness. However, OIRM has made
limited progress towards establishing a fully operational data standards program. In addition,
OW's only progress has been the issuance of Order 7500.1A (October 1992), Minimum Set of
Data Elements for Ground Water Quality. OW has not developed Information Technology (IT)
plans, data standards or standard guidance that will facilitate data sharing or integration of data
across OW systems. Data standardization is key to the Agency's goal to integrate data across its
information systems. Without data standardization and integration, EPA will continue operating
stovepipe systems and may not have the necessary environmental data needed to monitor state
water programs. The OW's lack of progress toward standardization and integration can be
attributed, in part, to the absence of a formal Information Resource Management (IRM) structure.
It is critical that OW and OIRM management work together to establish a structure to share data
across environmental systems, programs and media. To pursue the Agency's Reinventing
Environmental Information (REI) reforms, the OW Senior Information Resource Management
Officer (SIRMO) function should to start acting as a central clearing house for information across
the program and be responsible for developing OW's strategic information resource plans.
PURPOSE
The objectives of our survey were to review the Office of Water's development of plans to
integrate automated water systems data, and to determine whether Federal, Agency and OW
guidance was provided to ensure use of standard data elements for sharing information. We also
reviewed OW activities to increase the public's access to key water data.
SCOPE
We conducted this survey in Washington, D.C. Headquarters from October 1996 to September
1997. We performed our audit in accordance with generally accepted Government Auditing
Standards. (1994 revision) issued by the Comptroller General of the United States, and included
such tests as necessary to complete our objectives. Our survey did not include an evaluation of
management controls as it relates to OW's FMFIA process.
-------
Office of Water Data Integration Efforts
METHODOLOGY
We started the analysis with system and data information from the 1992 OW Environmental and
Program Information Systems Compendium, which identified 20 key systems and information
files as critical to OW programs. We concentrated on the national systems which support either
Safe Drinking Water or Water Quality Programs. We interviewed key personnel and reviewed
system documents to summarize the purpose and operation of three OW mission-critical systems
-- Clean Water Needs Survey (CWNS), Storage and Retrieval System (STORET), and Safe
Drinking Water Information System (SDWIS). We also summarized the purpose and operation
of the Permit Compliance System (PCS), which is an Enforcement system under the Office for
Enforcement and Compliance Assurance.
We developed a profile of OW IRM by interviewing management personnel and reviewing IRM
policy, budget, and planning documents for OW and OIRM. We interviewed key Agency
personnel and reviewed documentation associated with OW and OIRM projects to provide
environmental information to the public and integrate environmental data. We also documented
activities related to major systems modernization and attended related OW conferences. Further,
we documented major OW and OIRM systems projects and initiatives for public dissemination
of OW environmental data, such as Surf Your Watershed, Index Watershed Indicators and
Envirofacts. In addition, we summarized those problems and recommendations related to OW
data management from Federal and Agency workgroups or reported in notable audits conducted
by the Office of Inspector General (OIG) or General Accounting Office (GAO) during fiscal
vears 1995. 1996 and 1997.
BACKGROUND
The OIG initiated the survey to assist OW in implementing its vision of standardizing and
integrating water data. With the current structure of laws, regulations and program infrastructure,
OW believes it is time to view water systems as a whole (i.e., watershed), rather than on an
individual basis (e.g., lake or river). The OW information systems inventory, dated 1992,
identified 20 key systems and information files as critical to OW programs.
In December 1993, the National Performance Review (NPR) addressed the need for a seamless
interactive government, linking hundreds of systems across Federal, regional, state and local
government operations through a network of data highways. !n the same year, EPA
acknowledged the lack of an Agency data structure as an Agency weakness, based on an OIG
Report No. SI001'
-------
Office of Water Data Integration Efforts
report (# E1SKG3-15-0098-4400038) entitled ;iSpecial Review of EPA's Information Systems
Program," dated March 24, 1994. The OIG report identified more than 50 previously issued
OIG, GAO and Agency reports, as well as 19 testimonies which addressed Agency data
management problems. The OIG report identified IRM management structure, lack of a IRM
Strategic Plan, systems development, and data management as the principal areas of long-
standing IRM problems. Specifically, the OIG reported that:
Management did not treat information as a strategic resource nor IRM as a core function
and valuable tool to empower the public; and
Data sharing was critical to the Administrator's four top priorities which cut across all
environmental media, but the link between data sharing and these priorities was-not
widely recognized. EPA did not have an information data architecture, data standards, or
administrative structure to facilitate sharing data Agency-wide. Data quality problems
exist in many EPA systems because of changing data definitions, lack of data ownership,
inconsistent quality assurance processes in program offices, and EPA's inability to
compel the states to provide quality data.
In February 1995, Environmental Data Standards were recognized by the Intergovernmental Task
Force for Monitoring (ITFM), chaired by EPA and vice-chaired by the United States Geological
Service (USGS), as a government-wide concern. Also, the April 1995 National Academy of
Public Administration (NAPA) report to Congress, entitled "Setting Priorities, Getting Results,"
discussed the need to change the way EPA does business. NAPA pointed out that the single-
media program approach had resulted in more than 500 systems serving individual programmatic
needs. In addition, NAPA said that a particularly important problem was the lack of high quality
data on environmental conditions to measure progress toward environmental goals.
NAPA's 1997 report entitled "Resolving the Paradox of Environmental Protection" addressed
EPA's progress since the 1995 report. The report noted that EPA had been working to build a
base of environmental data to support performance-based management, but that progress was
slow. It further stated that the Agency still had not established institutions to ensure data was
reliable, objective, credible, and consistent across programs and media. NAPA's progress report
also stated EPA's environmental data systems were not adequate to measure environmental
trends because of data management problems. NAPA noted that, in 1997, EPA's National
Advisory Council on Environmental Policy and Technology warned EPA that, absent improved
information management, the success of regulatory flexibility, place-based approaches and cross-
Repori Ho. 8100/7'
-------
Office of Water Data Integration Efforts
media approaches could be at risk, ft proposed simplified reporting and use of environmental
data by establishing data standards for EPA's programs and a bureau of environmental statistics.
As a result of the REI initiative, Agency management is developing a plan for standardizing data
and accepting a stewardship role over environmental information. On July 21, 1997, at a
Common Sense Initiative (CSI) Council Meeting, EPA's Administrator and Deputy
Administrator announced EPA would pursue three important information management reforms:
establishing key data standards, providing universal voluntary access to electronic reporting, and
implementing these data standards and electronic reporting reforms through the One Stop
program. The resulting REI Action Plan, dated February 4, 1998. identified thirteen national
systems which are expected to incorporate EPA's data standards, policies and protocols.
In March 1996, EPA's Administrator asked the National Advisory Council for Environmental
Policy and Technology (NACEPT) to review EPA's information requirements and processes. In
January 1998, NACEPT issued their report, Managing Information As A Strategic Resource,
citing the following findings:
1. EPA cannot develop the required information resources unless:
The agency establishes information as a strategic tool,
The public is provided information to monitor performance.
Industry has information needed to develop prevention options, and
All stakeholders are provided required information.
2. Current EPA systems for the most part were not designed to support and do not provide
place-based, multi-media, and cross-media approaches to environmental protection.
Additional detailed background information is presented in Appendix II to this survey report,
PRIOR AUDIT COVERAGE
The following reports do not represent a complete listing of reports concerning EPA's Water
Program, but rather feature recent, key reports dealing with the questionable quality of collected
water data and the need for automated data management systems to ensure more consistent and
efficient data management.
Report No. 810017',
-------
Office of Water Data Integration Efforts
In GAO/RCED report number 96-42, dated January 23, 1996, "Water Pollution - Differences
Among the States in Issuing Permits Limiting the Discharge of Pollutants," GAO stated that
permit limits for controlling pollution within watersheds differ widely because of differences in
the (1) states' water quality standards and (2) implementation policies that come into play when
the permitting authorities "translate" general water quality standards into limits for specific
facilities in specific locations. In response, EPA management said they plan to enhance reviews
of state implementation policies.
DIG report number E1HWF6-07-0017-6100312, dated September 30, 1996, entitled "Region 7s
Efforts to Address Water Pollution From Livestock Waste," identified the need to use state water
quality information for controlling waste through permits in delegated programs. The report
concluded that EPA could support the states by analyzing trends to demonstrate the benefits,
effects and scope of pollution.
Likewise, OIG report number E1HWF4-07-0036-5100226, dated March 24, 1995, entitled
"Region 7 and States Improved Drinking Water Programs Through Alternative Measures,"
identified that states needed automated data management systems for more consistent and
efficient data management. EPA was relying on manual or partially automated systems which
were inefficient and vulnerable to inaccuracies. Subsequently, several OW systems were
modernized with the help of state participation, but OW still cannot mandate use of these
systems.
In OIG report number E1HWE5-23-OOOI-5100516, entitled "EPA Procedures to Ensure
Drinking Water Data Integrity," dated September 29, 1995, we projected that about 12 percent of
nationwide public water services reported erroneous data one or more times from 1991 through
1994. Small public water services most often reported erroneous data and about 58 percent of
the erroneous data cases involved invalid data, rather than data which might have been
elaborately falsified.
In addition, GAO/RCED report number 94-9, dated February 17, 1994, entitled "Water
Pollution: Poor Quality Assurance and Limited Pollutant Coverage Undermine EPA's Control of
Toxic Substances," found that some of the information used to analytically support in-stream
water quality for toxic pollutants may be of doubtful quality, thereby raising questions about the
effectiveness of these activities in controlling toxic pollutants. At that time, GAO also found that
the permit process did not limit the vast majority of toxins being discharged from the nation's
Report No. 810017'
-------
Office of Water Data Integration Efforts
factories and sewage treatment plants. GAO's audit results showed that five of the seven core
activities did not measure up to EPA's enunciated information quality assurance policies and
principles. GAO's findings raise questions about the quality of information generated and used
within these activities. Many of the EPA interviewees noted that data on in-stream water quality
for toxic pollutants was generally lacking. GAO was able to confirm the accuracy of these
responses through a review of the data in STORET, EPA's principal repository of water quality
data. For a group of streams around the nation on which large manufacturing facilities are
situated, GAO found that only 9 percent had any in-stream data for organic toxic pollutants or
pesticides entered in STORET since 1988.
RESULTS IN BRIEF
Our survey indicated OW has begun addressing water data integration, but progress is slow and a
more focused effort is needed. Except for Order 7500.1 A (October 1992), Minimum Set of Data
Elements for Ground Water Quality, no IT plans, data standards, or guidance to facilitate the
sharing or integration of data across OW mission-critical systems have been developed. Without
standardization and integration, EPA will continue operating stovepipe systems and may not
have the necessary environmental data needed to monitor state water programs. Both OIRM and
OW's lack of progress toward standardization and integration can be attributed, in part, to the
absence of a formal Information Resources Management structure.
The O W is starting to address the cross media, data standards and sharing of water data in the
following ways:
OW is involved on an intergovernmental level with the ITFM's successor organization,
the National Water-Quality Monitoring Council, which is working oh implementing
ITFM recommendations for Federal agencies.
The recent national OW systems modernization efforts are moving in the right direction.
For example, OW is conducting meetings and conferences to determine users' needs
(including the States). SDWIS, a mission-critical system, has begun to address data
standards for reporting for that program. Also, STORET data has been entered into the
Environmental Data Registry (EDR).
Report No. 8100177
-------
Office of Water Data Integration Efforts
At the time of our review, OW had numerous, division-level projects to (1) increase
public access to key water data through the Internet, (2) organize water data by
watershed and (3) establish environmental indicators. OIRM is providing more OW
data through EPA's Envirofacts Warehouse, a project to provide data through EPA's
Internet home page. Several of these pilots are now operational sites providing data to
the public.
The OW should support EPA's REI initiatives, which will require aggressively addressing
environmental data standards and integration. In addition, OW should establish long-range
strategic plans with measurable goals over the next three to five years for its major information
systems. The present management structure allows each program, division, region and state to
establish different data definitions. Therefore, to accomplish the REI plan, OW should centralize
its IRM structure and work with OIRM to adopt an environmental data standardization process
that will be mandatory for national water systems. On July 21, 1997. at a CSI Council Meeting.
the Administrator and Deputy Administrator announced that EPA would pursue three important
information management reforms: (1) establishing key data standards, (2) providing universal
voluntary access to electronic reporting, and (3) implementing these data standards and electronic
reporting reforms through the One Stop program. To pursue the reforms, the OW SIRMO
function should start acting as a clearing house for information across the program. Then the
SIRMO would be in a position to reduce duplication of systems, institutionalizing the lessons
learned in pilots and enhancing the mission-critical national systems through action plans,
environmental data standards or environmental data guidance.
Report No. 8100H'
-------
Office of Water Data Integration Efforts
FINDINGS AND RECOMMENDATIONS
Water Data Integration Needs A Centralized Focus
Despite NPR goals for establishing a seamless interactive government, ITMF's recognizing the
need for water data standards, and the OMB requirement for the Agency to address information
system standard architecture in its budget, OW has been slow to address water data integration
and a more focused effort is needed to meld various OW activities and systems. As of yet, OW
has not developed data standards or standard guidance that will facilitate data sharing or
integration of data across OW systems. The OW is modernizing its mission-critical systems:
SDWIS. CWNS, and STORET. However, each OW system is being modernized independently
of the others. Thus, the modernization efforts are not resulting in standard water data elements as
envisioned by the ITFM. Although the modernized systems will be more useful, relational-type
databases, they will continue to be stovepipe systems that will not standardize or share their data.
White OIRM management acknowledged data management as an agency-level weakness in
1994, progress has been slow in establishing an operational data standards process and data
standards. In September 1997, the Agency took a first step toward IT planning by submitting
proposals for EPA's IT Investment Report to OMB. Also, at the time of our audit, OW and
OIRM had initiated pilot projects to: (1) provide the public with OW data from the national
water systems, (2) organize water data by watershed, and (3) establish environmental indicators
that are being provided to the public through the Internet. OIRM is also piloting a data standards
process in conjunction with the EDR, a data standards registration tool, for the six standards
identified in the REI Action Plan. Data standards established to date include standards for
location, facility IDs, and a four-digit year date.
OIRM management is planning to establish a standard Agency architecture and formally adopt a
new data standards process that uses the EDR. These particular actions are currently scheduled
for completion by March 1999, but it appears that it will be at least fiscal 2003 before minimal
data standards are actually implemented. The REI Action Plan, dated February 4, 1998,
established implementation milestones for the next three to five years for incorporation of the
Agency data standards. For example, an electronic reporting policy will be published by the end
of fiscal 1999 for the systems identified in the REI Action Plan, and these systems will use
electronic reporting standards by fiscal 2003. Therefore, OW should incorporate electronic
reporting standards resulting from the REI effort for SDWIS and STORET by fiscal 2003. The
OW should also coordinate with Office of Enforcement and Compliance Assurance on
incorporating these standards into PCS. Without standardization and integration, EPA will
continue operating stovepipe systems and may not have the necessary environmental data needed
to monitor state water programs. Standardizing data elements would ease integration of systems
8
Report No. 8100177
-------
Office of Water Data Integration Efforts
which, in turn, could foster cost-effective sharing of available water information required for
effective decision-making and management of environmental protection.
The OW's lack of progress toward standardization and integration resulted primarily from the
absence of a formal IRM structure and lack of formal guidance from EPA OIRM.
Currently, the OW does not have a. formal IRM structure to provide oversight and
approval of water systems development. At the time of our survey, the OW used a
decentralized team management process to address its information requirements and had
not tasked anyone with developing a standard architecture. In addition, the SIRMO
function had no control over OW systems projects, leaving systems development to be
managed at a director or branch manager level. Since our audit fieldwork, OW has
established an office-wide IRM committee as the focal point for increasing efforts to
share and integrate data across OW programs. However, the committee is operating
without a charter which means they have no binding authority to make IRM decisions
across OW programs or accountability for those decisions. It is critical that OW
management establishes a structure to share data across environmental programs and
media. To pursue the REI reforms, the OW SIRMO function should start acting as a
clearing house for information across the program.
Although EPA's OIRM has taken preliminary steps to set up an EDR, management has
not yet established formal policy or procedural guidance for either the EDR or data
standardization. Also, EPA's OIRM has not required programs to address environmental
data element standards as part of new systems development, although the IT Investment -
Draft Data Call dated February 24. 1998, requires managers to identify whether or not
they are complying with Agency data standards when requesting ESC review and
approval. Upon reviewing the number and nature of Agency initiatives since this
weakness was first acknowledged, it appears that the Agency predominantly focused its
resources on projects to provide public access to environmental information, rather than
investing effort in projects to increase cross-media integration of environmental data.
Federal And Agency IT Policies And Procedures
OMB Memorandum 97-16, dated June 18, 1997, requires Federal agencies to develop and
implement Information Technology Architectures (ITA). The ITA describes the relationships
among the work the agency does, the information the agency uses, and the agency's IT needs. It
includes standards to guide the design of new systems. An ITA makes it easier to share
information internally and to reduce the number of information systems that perform similar
functions. The ITA provides the technology vision to guide resource decisions that reduce costs
Report No. 8100177
-------
Office of Water Data Integration Efforts
and improve mission performance. In addition, OMB Memorandum 97-02, Funding Information
Systems Investments, dated October 25, 1996, requires Agency investments in major information
systems to be consistent with Federal, agency, and bureau Information Technology Architectures
(ITAs). Also, the Clinger-Cohen Act of 1996, (Public Law 104-106), assigns the Chief
Information Officer (CIO) the responsibility of developing, maintaining, and facilitating the
implementation of the ITA. Agencies are to be prepared to indicate the status of the
development, implementation, and maintenance of the agency ITA during the formulation of the
fiscal 1999 President's budget. This is a new requirement which EPA must comply with starting
this year.
OW INFORMATION TECHNOLOGY PROJECTS
ESTIMATED SPENDING THROUGH FISCAL 2003
LIFE CYCLE COSTS '
$30,553,000
$61.710,000
$11,233,000
$18,560,000
$12,710,000
NAME
Storage and Retrieval of Water Quality Information
Safe Drinking Water Information System Modernization -
Surf Your Watershed
Information Collection Rule
The Index of Watershed Indicators - formerly NWAP
The preceding chart identifies major capital investments for OW major systems, initiatives, and
IT infrastructures with Life Cycle Costs more than $ 5 million, as reported to OMB in September
1997. Major capital IT investments are being managed under Circular A-l 1, Part 3; and the
Information Technology Management Reform Act (ITMRA) of 1996. EPA's Directive 2100,
Chapter 17, establishes policies and procedures for management of systems enhancements and
modernizations. The Directive mandates EPA's Assistant Administrators to directly approve .
investment of information resources for major systems costing $1 million a year or having a total
life-cycle cost of $10 million or considered mission-critical. In April 1997, the CIO also
1 Life cycle costs contain estimates of all operational and developmental costs, including
full time equivalents for personnel.
2 Costs include 'SIDWIS/FED, SDWIS/State, and National Contaminants Occurrence
Data Base.
10
Report No. 8100177
-------
Office of Water Data Integration Efforts
established an Investment Review process, under ITMRA, in which the Executive Steering
Committee performs reviews of initiatives or infrastructure greater than $1 million a year.
Program guidance includes EPA Order 7500.1 A (October 1992). Minimum Set of Elements for
Ground Water Quality, which established a policy for a minimum set of data elements for
ground water quality to be collected and managed by the EPA ground water data collection
activities. A second EPA Order 5360.1 (April 1984), Policy and Program Requirements to
Implement the Mandatory Quality Assurance Program, established EPA program quality
assurance requirements and policies for all environmentally-related measurements performed by
or for EPA.
Need To. Establish Formal IRM Structure For Oversight
Of Information Management
The recently implemented REI Action Plan affects all program offices and requires a more robust
Agency and OW IRM oversight structure. This will require clear lines of authority'within OW to
establish policy, schedule changes, and allocate resources. In addition^ OW should work with
OIRM to establish a standard architecture and water data standards for EPA. Without
standardization and integration. EPA will continue operating fragmented systems and may not
have the necessary environmental data needed to adequately monitor state water programs.
Standardizing data elements would be a first step to the integration of OW systems and
environmental data required for effective decision-making and management purposes.
Similar to attributing EPA's slow progress towards data standardization and integration to the
absence of a formal AA-level IRM control structure and the lack of USGS or OIRM guidance,'
OW?s lack of progress can be ascribed to its treatment of IRM as a support structure and the
absence of formal O W IRM guidance. At the time of our survey, the OW used a decentralized
team management process to address its information requirements and had not tasked anyone
with developing a standard architecture. And while the SIRMO reported to the AA, that
individual had no direct oversight of operations or budgets for OW systems projects, initiatives
or infrastructure, leaving systems development to be managed at a director or branch manager
level. Since our audit fieldwork. OW has established an office-wide IRM committee as the focal
point for increasing efforts to share and integrate data across OW programs. However, the
committee is operating without a charter which means they have no binding authority to make
IRM decisions across OW programs or accountability for those decisions.
Until recently, EPA treated information management as a support function rather than a central
tool to manage its programs and inform the public. Some recent management decisions have
sought to recognize the contribution information makes to environmental protection and elevate
11
Report A'o. SlOOl"
-------
Office of Water Data Integration Efforts
the information management function. For example, in 1995. the Executive Steering Committee
was established to manage cross-Agency Information Resource needs. In addition, on February
27, 1997, the Agency instituted a Center for Environmental Information and Statistics (CEIS) to
centralize EPA's information management efforts and provide a strategic focus to EPA's
information collection, assessment and access efforts. CEIS formally began operations in
January 1998. OIRM has also taken preliminary steps to set up an EDR, but it has not yet
established formal policies or procedures necessary for the registry's operation or data
standardization program. Additionally, EPA's OIRM has not required programs to address
environmental data element standards as part of new systems development, although the IT
Investment - Data Call dated February 24, 1998, does request managers to report on compliance
with Agency data standards for ESC review of major projects.
OIRM established an Information Management/Data Administration group in 1992, to define
and establish data management policies like Facility ID and Locational Data Policy. In addition.
EPA management formally acknowledged Data Management issues in fiscal 1994, as an
Agency-level weakness in their annual FMFIA report. The original targeted corrective action
date was December 1998. However, progress has been slow and only recently did OIRM start to
put into place formal Agency processes to establish data standards. To date, standards
established or in-process include standards for a four-digit year date, location (i.e., latitude and
longitude), facility ID, Biological Taxonomy, Chemical ID, and Industrial Classification
(SIC/NAICS) codes. EPA's date for establishing environmental data standards has slipped twice
since fiscal 1994, and is currently set for March 1999. The REI plan calls for finalizing a total of
six standards by 1999. Considering the focus of ongoing Agency initiatives, it appears that the
slippage is a result of the Agency targeting projects to provide "public access" to environmental
information, rather than investing resources in projects to improve cross-media integration of -
environmental data.
In our opinion, without a formal IRM structure and direction from EPA OIRM, OW will not
achieve standardization of data elements and integration of systems needed to manage based on
water bodies. The shifting of program IRM resources to a centralized basis should allow O W to
address data standardization and system integration while reducing division-level resources
needed for national systems.
Federal And State Agencies Have Not Made Standard
Architectures And Data A Priority
The creation of standard architectures and data are not a government-wide priority for all Federal
and State water programs. USGS has not developed Federal data standards for collecting water
data by Federal Organizations, and there are few interagency agreements between EPA and other
12
Report No. 8100177
-------
?r*"? r>
Office of Water Data Integration Efforts
Federal Organizations currently providing data for EPA's programs. In addition, there is much
missing data that is not available from other Federal Organizations or States.
In 1995, the ITFM identified the need to address two data related areas (1) information
automation, accessibility and utility and (2) data quality. In general, the ITFM noted the need to
develop additional tools to facilitate information searches and provide capability for retrievals
across data bases. One such tool was a set of minimum data elements for sharing existing data.
However, after about three years, there are still no formal intergovernmental water standards in
operation today, because a formal coordination structure is just starting to operate and the study's
recommended actions were "voluntary."
The recently established Federal CIO Council formed a subcommittee on inter-operability that
plans to study interagency integration issues. Also, in October 1997, the CIO council published a
reference guide on Information Systems Interoperability. Although State and Federal
Organizations are modernizing their computing platforms, installing more open systems and
upgrading their networks, they have not necessarily made integration with each other a priority.
For example, the CIO for California said "We're dealing with things on a state level, and a lot of
the time we don't have the luxury to deal with projects an intergovernmental way." The primary
reason data does not match from state to state is that one state may call something a road, while
some other state calls it a highway or a byway. Frequently, these definitions are legal ones.
which means new laws or regulations are needed to create common definition requirements.
OW Information Systems Will Continue To Be Fragmented
With Inadequate And Unreliable Data
Under existing circumstances, OW will continue operating fragmented information systems with
inadequate and unreliable environmental data and, as a result, will not effectively manage or
monitor state water programs. The quality of OW's systems data is a recognized long-standing
problem, reported by NAPA, NACEPT, GAO and the OIG. From 1990 through 1994; numerous
OIG and GAO reports, as well as Agency reports and testimonies addressed EPA's information
systems data problems. In addition, NAPA's 1997 report stated that data available to EPA is
incomplete, fragmented, unreliable and difficult to use. In January 1998, NACEPT also reported
that EPA should establish environmental information as a strategic tool for all stakeholders.
Since 1994, the OIG and GAO issued additional reports addressing water program data
management by states and regions.
Attainment of the "seamless, interactive" government imagined by the NPR is threatened by the
lack, so far, of a comprehensive strategy for electronically stitching together the hundreds of
Federal organizations and state systems. There is much missing data that is not available from
other Federal organizations or states. Also, EPA is receiving data of unknown quality from other
13
Report No. 8100177
-------
Office of Water Data Integration Efforts
Federal organizations and putting it in our national systems to monitor water quality. In 1992,
the USGS was assigned responsibility to develop mandatory Federal data standards for collecting
water data by Federal organizations. Only recently, the National Water-Quality Monitoring
Council was established with procedures to coordinate, provide guidance and technical support
for the implementation of the recommendations presented in the Strategy for Improving
Water-Quality Monitoring by U. S. government organizations and the private sector. However,
the Council has yet to establish Federal data standards for water programs. Therefore, EPA's
ability to obtain Federal data needed to monitor water quality is still heavily dependent on
informal agreements with numerous organizations.
Recommendations
Due to the complex nature of the.issues, it is our opinion that both OW and OIRM need to
implement corrective actions to effectively address these weaknesses. We recommend that the
Assistant Administrator for Water:
I. Establish a centralized IRM management organization by a formal charter which assigns
responsibility for oversight and budget of programs' IT capital planning investments,
operation of major systems, approval and oversight of initiatives, and development of a
standard OW Infrastructure.
2. Establish an architecture and coordinate data management activities with USGS for
developing Federal water data standards, adhering to technology standards and a common
design, and providing guidance for OW programs.
3. Aggressively support the implementation of the REI Action Plan by working with OIRM to
develop EPA-level environmental data standards.
In addition, we recommend that the Director for Information Resource Management:
4. Formally adopt policies and procedures in Agency Directive 2100 to support an Agency data
standards program using the EDR as EPA's central repository for publishing and recording
data standards.
14
Report No. 8100177
-------
Office of Water Data Integration Efforts
AGENCY COMMENTS
In responding to the draft report, the Acting Director for Information Resources Management and
the Acting Director for Policy and Resources Management, Office of Water, generally agreed
with the recommendations. Further, they acknowledged that much remains to be done and stated
that both offices are committed to continuing and strengthening their efforts. OW has
established an office-wide IRM committee as the focal point for increasing efforts to share and
integrate data across OW programs. OW also stated that they have been actively engaged in
coordinating data management activities with USGS and other Federal Agencies. Currently. OW
is involved in the National .Water Quality Monitoring Council, which OW co-chairs with USGS.
and the Federal Geographic Data Committee on the River Reach File, Version 3. OW has also
been active in several committees implementing the REI initiative. OIRM management stated
that they were committed to establishing policies and procedures to ensure an operational
program, and plan to incorporate data standards into the new IT investment review process and
other Agency business practices. OIRM management stated that they plan to use the EDR as the
. primary tool to publish data standards. There was a general concern that the OIG implied OW
should move ahead on its own to establish water data standards for its systems and. therefore, the
joint response suggested revisions to correct this impression.
OIG EVALUATION
We agree that the Agency's proposed actions generally address our findings and
recommendations, and we modified recommendations 1 and 4 to clarify their intent. In
recommendation 1. we believe the Office of Water should establish a formal charter for their
new IRM committee. At present, the committee is operating without a charter which, in our
opinion, is required to establish (1) a formal decision process, (2) authority for the committee to
make IRM decisions that are binding across the OW programs and (3) accountability. We would
anticipate that one of the first products this committee would establish would be a plan
addressing significant OW IRM projects, including each project's objective(s), budget and
expected milestones. In recommendation 4, we also agreed with some suggested wording
changes to clarify the recommendation. We believe OIRM should formally adopt policies and
procedures for data management by modifying OIRM's Directive 2100. The March 1999
FMFIA schedule for completing these actions is ambitious, but management assured us that it
will be completed on schedule. Finally, based on management's response, the OIG made some
changes to the final report which should clarify our desire for OW to work with both OIRM and
other Federal programs to develop OW data standards.
15
Report No. SI00177
-------
Office of Water Data Integration Efforts
THIS PAGE INTENTIONALLY LEFT BLANK
16
Report No. 81001'
-------
Office of Water Data integration Efforts
APPENDIX I
Page 1 of I
ABBREVIATIONS
CEIS
CIO
CSI
CWNS
EDR
FMFIA
GAO
IRM
IT
ITA
ITFM
ITMRA
NAPA
NACEPT
NPR
OIG
OIRM
OW
PCS
REI
SDWIS
SIRMO
STORE!
USGS
Center for Environmental Information and Statistics
Chief Information Officer
Common Sense Initiative
Clean Water Needs Survey
Environmental Data Registry
Federal Managers' Financial Integrity Act
General Accounting Office
Information Resources Management
Information Technology
Information Technology Architecture
Intergovernmental Task Force for Monitoring
Information Technology Management Reform Act
National Academy of Public Administration
National Advisory Council for Environmental Policy and Technology
National Performance Review
Office of Inspector General
Office of Information Resources Management
Office of Water
Permit Compliance System
Reinventing Environmental Information
Safe Drinking Water Information System
Senior Information Resource Management Officer
Storage and Retrieval System
United States Geological Service
17
Report No. 8100177
-------
Office of Water Data Integration Efforts
THIS PAGE INTENTIONALLY LEFT BLANK
18
Report No. 81001'
-------
Office of Water Data Integration Efforts
APPENDIX II
Page 1 of 5
FEDERAL DATA MANAGEMENT
National Performance Review
The 1993 NPR report on EPA identified the need for pilot projects to test (1) the usefulness of
and access to environmental information and (2) access and exchange of environmental
information among Federal, states, tribal, local and foreign governments. In addition, the report
noted that EPA must improve the data quality and analysis for regulations. In 1993, the NPR
addressed the need for a seamless interactive government, linking hundreds of systems across
Federal, regional, state and local government operations through a network of data highways.
Principal Areas of Long-standing IRM Problems
OIRM acknowledged the lack of an Agency data structure as an Agency weakness based on an
OIG report number E1SKG3-15-0098-4400038 entitled "Special Review of EPA's Information
Systems Program," dated March 24, 1994. The report identified more than 50 OIG, GAO and
Agency reports, as well as 19 testimonies addressing Agency problems over a four year period.
It identified four principal areas of long-standing IRM problems:
1. IRM management and organizational structure did not understand how critical IRM was
to accomplishing environmental protection. Management did not treat information as a
strategic resource, IRM as a core function and information as a valuable tool to
empower the public.
2. Resource planning and performance management needed to be integrated into EPA's
budgeting process.
3. Information systems development did not adequately consider users and customers.
4. Data sharing was critical to the Administrator's four top priorities which cut across all
environmental media, but the link between data sharing and these priorities was not
widely recognized. EPA did not have an information data architecture, data standards,
or administrative structure to facilitate sharing data Agency-wide. Data quality
problems exist in many EPA systems because of changing data definitions, lack of data
ownership, inconsistent quality assurance processes in program offices, and EPA's
inability to compel the states to provide quality data.
EPA still does not have an information management structure which includes a data architecture,
data standards and supporting administrative functions that would permit sharing of
19
Report No. SI00177
-------
Office of Water Data Integration Efforts
APPENDIX II
Page 2 of5
environmental data Agency-wide. The corrective action date has slipped twice since fiscal 1994,
and the Acting Chief Financial Officer's FMFIA report, dated December 12, 1997, now indicates
that corrective action will be completed by March 1999.
Intergovernmental Task Force for Monitoring
In February 1995, Environmental Data Standards was recognized as a government-wide concern
by an ITFM report which recommended a strategy for nationwide, integrated, voluntary water-
quality monitoring. The ITFM, chaired by EPA and vice-chaired by the USGS, declared water
data standards as necessary for Federal governmental programs. The 1995 ITFM study identified
the need to address two data-related areas: (1) information automation, accessibility and utility,
and (2) data quality. In general, the ITFM noted the need to develop additional tools to facilitate
information searches and provide capability for retrievals across data bases. One such tool is a
set of minimum data elements for sharing existing data. The study stated that EPA should
voluntarily incorporate the following during modernization of its old systems or when new
systems are being developed:
Common data-element definitions and formats.
An expanded set of recommended data elements or qualifiers (in addition to the
minimum data elements) to facilitate the sharing and exchange of information.
Common reference tables, such as taxonomic and hydrologic unit codes, and River
Reach File 3 codes.
Metadata standards to describe the content, quality, condition, and other
characteristics of data. Metadata helps secondary users to judge whether the data
would be useful for other application.
Facilitate the sharing of water-quality information that would be useful to '
secondary users, but that currently is not readily available. For example, major
public-water suppliers have offered to share such information holdings.
Share, and where advantageous, jointly maintain ancillary data sets that are widely
used for water-quality purposes, such as land use, land cover, demographics, and
water use. Working with the Federal Geographic Data Committee and other
groups, use standard data sets when they are available. An example would be the
River Reach File that is being jointly developed and adapted as part of the FGDC's
National Spatial Data Infrastructure.
Use Internet or other widely-recognized standard communications and access
systems when they are available.
20
Report iVo. 810017'
-------
Office of Water Data Integration Efforts
APPENDIX II
Page 3 of5
EPA should also voluntarily establish, for all environmental water monitoring programs, data-
quality objectives to identify the precision and accuracy of data needed to achieve monitoring
goal(s).
NAPA Studies
NAPA's report to Congress in April 1995, entitled "Setting Priorities, Getting Results,"
discussed the need to change the way EPA does business in six areas. One of these areas
addressed how EPA was allocating its resources and establishing priorities to accomplish its
mission. NAPA recommended that EPA develop strong central management systems to manage
its activities, and refine and expand the use of risk analysis and cost benefit analysis to make
decisions. The report also recommended establishing an independent statistics organization to
assess data needs, assemble data, analyze data, and disseminate information. It pointed out that
the single-media program approach had resulted in more than 500 systems serving individual
programmatic needs. In addition, NAPA said that a particularly important problem was the lack
of high quality data on environmental conditions to measure progress toward environmental
goals.
In September 1997, NAPA reported to Congress on Resolving the Paradox of Environmental
Protection, which addressed EPA's progress since 1995. The report noted that EPA has been
working to build a base of environmental data to support performance based management, but
that progress is slow. It further stated that the Agency still has not established institutions to
ensure data is reliable, objective, credible, and consistent across programs and media. NAPA's
progress report also stated EPA's environmental data systems are not adequate to measure
environmental trends because:
Data available to EPA is incomplete.
Data is fragmented and difficult to use. The systems were designed by separate program
offices to meet their individual needs. EPA relies on other agencies to collect data and
those agencies have their own standards. In fact. State and local data systems contain
three or four times more data than is reported to EPA. EPA has not exerted much quality
control over the data being collected.
Until recently, EPA treated information management as a support function rather than a
central tool for managing and interacting with the public.
NAPA noted that, in 1997, EPA's National Advisory Council on Environmental Policy and
Technology warned EPA that, absent improved information management, the success of
21
Report ;Vo. 81001:
-------
Office of Water Data Integration Efforts
APPENDIX II
Page 4 of 5
regulatory flexibility, place-based approaches and cross-media approaches could be at risk. It
criticized the Agency's Executive Steering Committee as ineffective and again recommended
creating a bureau for Environmental Statistics. It also proposed simplified reporting and use of
environmental data by establishing data standards for EPA's programs and a bureau of
environmental statistics.
EPA's Common Sense Initiative
On July 21, 1997, at a CSI Council Meeting, the Administrator and Deputy Administrator
announced EPA would pursue three important information management reforms: establishing
key data standards, providing universal voluntary access to electronic reporting, and
implementing these data standards and electronic reporting reforms through the One Stop
program. The resulting Action Plan for Reinventing Environmental Information, dated February
4, 1998. identified thirteen national systems which are expected to incorporate data-standards,
policies and protocols. OW'S STORET and SDWIS are among the thirteen systems targeted in
that plan. Each system will incorporate the standards within three years. Also, as each new data
and reporting standard becomes ready for implementation, each national system has two years in
which to incorporate it either through retrofitting existing systems or including the standard in
system re-engineering efforts. The present management structure allows each program, division,
region and state to establish different data definitions. Therefore, to accomplish the REI plan,
OW should centralize information resources management and work with OIRM to adopt an
environmental data standardization process which will be mandatory in all OW mission-critical
systems. This will then support goals for (1) integration of water data for the general public, .
(2) monitoring the states' water programs, and (3) managing the national water programs.
NACEPT Report
NACEPT is a public advisory committee chartered on July 7, 1988 that provides
recommendations and advice to EPA's Administrator. NACEPT was requested in March 1996
to review EPA's information requirements and processes and make recommendations to make
EPA's.information resources support Community-Based Environmental Protection and other
initiatives like the Common Sense Initiative. In January 1998, NACEPT issued a report.
Managing Information As A Strategic Resource. As a result the Committee offered the
following findings:
1. EPA is moving toward defining the information needed, but it can't develop the required
information resources unless:
Report No. SI00177
-------
Office of Water Data Integration Efforts
APPENDIX II
Page 5 of5
EPA establishes information as a strategic tool,
The public is provided information to monitor performance,
Industry has information to develop prevention options, and
AH stakeholders are provided required information.
2. Current EPA systems for the most part were not designed to support and do not provide place-
based, multi-media, and cross-media approaches to environmental protection. Thus, the systems
do not support:
Informed decision making,
Ensure accountability, or
Document results and achievements.
The committee also recommended actions and time frames to go beyond EPA's current
approach of pollution control oriented systems to a place-based, cooperative approach with
stakeholders. The goal is to make IRM a strategic tool that is integrated into accomplishing its
mission. Some of the actions included:
1. Information Policy Leadership
Appoint a full time CIO;
Designate one official over information to use it as a strategic resource;
Make permanent selections for all senior IRM positions currently filled in a acting capacity:
Agency should expand the role of the Executive Steering Committee for IRM.
2. Data Integration (develop a plan to integrate systems and resolve Federal-State barriers.)
3. Data Accuracy (develop an implementation plan to establish Agency system architecture and
address improving the precesses record and correct data).
4. Public Access (immediately take various actions to improve access and establishing a Public
Access Program)
5. Stakeholder Involvement (immediately set up a Information Users Group).
23
Report No. 8100177
-------
Office of Water Data Integration Efforts
THIS PAGE INTENTIONALLY LEFT BLANK
24
Report No. 8100177
-------
Office of Water Data Integration Efforts
APPENDIX III
Page 1 of2
May?, 1998
MEMORANDUM
SUBJECT:
FROM:
TO:
OW and OIRM Response on the Draft Report on the Office of Water's Data Integration
Efforts Audit Report No. E1NWG6-15-0001
Diane Regas, Acting Director /signed/
Policy and Resource Management Office, Office of Water (4101)
Mark Day, Acting Director /signed/
Office of Information Resources Management (3401)
Michael D. Simmons
Deputy Assistant Inspector General for Internal Audits (2421)
We have reviewed the draft report and generally agree with the recommendations. While both of
our offices have been actively working, in some cases for years, toward these goals, we acknowledge that
much remains to be done and are committed to continuing and strengthening our efforts.
Recommendation I. Establish a centralized IRM management organization. Within the last year,
OW established an office-wide senior IRM committee consisting of Deputy Office Directors, the Deputy
Assistant Administrator, the Acting Director of the Policy and Resource Management Office (PRMO),
the SIRMO and IRM staff from the Immediate Office. This committee has been involved in guiding
development of OW's IRM strategic plan; discussion of cross-office information initiatives; indexing of
programmatic data to the River Reach Fi.le for sharing and use of OW data in a geographical, or place-
based, context; and common OW budget initiatives for enhancing our program's effectiveness though use
of common data elements. This group is the logical focal point for increasing OW efforts to share and
integrate our data across our offices.
Recommendation 2. Establish an architecture and coordinate data management activities with
USGS. OW has been actively engaged in coordinating data management activities with USGS and other
Federal and State agencies through the National Water Quality Monitoring Council, which OW co-chairs
with USGS. A key goal of this organization is to develop consistent standards for data on water
monitoring. We also participate in the Federal Geographic Data Committee and work collaboratively
with USGS on establishment and implementation of the River Reach File, Version 3 (RF3). The OW
IRM committee is considering a long-term project to index water data to the RF3 to enable better spatial
data analysis and is examining resource availability' for this effort.
25
Report A'o. 81001''-
-------
Office of Water Data Integration Efforts
APPENDIX III
Page 2 of2
Recommendation 3. .Aggressivelysupport the implementation of the REJ plan. OW has also been
actively engaged in the Reinventing Environmental Information initiative. OW was represented on the
REI subcommittee by Diane Regas, Acting Director, PRMO, and was also represented on the various
REl-related suborganizations. OW is represented on the REI National Systems Board and the data and
EDI standards workgroups. We have planned for OW cross-office representation on these workgroups to
ensure that all offices are aware of and participating in standards that affect their data and systems.
Senior OW management is kept apprized of REI progress at meetings of the OW IRM committee. We
have committed to REI and will continue to work for its effective implementation.
Recommendation 4. Formally adopt policies and procedures establishing the EDR as EPA's data
standards and data architecture. OIRM is committed to establishing a strong Data Standardization
Program under the Agency's Reinventing Environmental Information (REI) Plan. That commitment
carries with it the requirement to put into place the necessary administrative structure, i.e., the policies
and procedures, to ensure an operational program. Our action plan includes plans to revise the IRM Data
Standards Policy (Chapter 5 of the IRM Policy Manual) and to create procedural pieces to explain how
the standards program will operate. New policies are being considered to clearly define the role and
responsibilities of the Data Steward and the Data Registrar, two critical elements to a successful
standards program. In addition, OIRM is planning to use the EDR web site as the primary tool for
publishing policies and procedures pertaining to the Agency's standards program to ensure their ongoing
visibility. OIRM plans to move the data standards mandated by the REI Plan through the new standards
setting processes. Finally, OIRM is incorporating data standards into routine agency business practices
and reviews such as the IT investment review mandated by the Clinger-Cohen Act. For the first time this
spring, Agency program offices must include in their IT funding requests descriptions of how their
information systems will comply with agency data standards.
In reviewing the draft audit, there was a general concern about inconsistencies in the suggestions on
how the Agency should operate its data standards program. Some language in the audit recognizes that
the establishment and use of data standards is an Agency-wide responsibility, other sections of the audit
imply that OW should move ahead on its own to establish water data standards. By recommending that
OW establish data standards for its systems without at the same time recognizing the Agency-wide
implication, we risk continuing the stovepipe approach to information system management. Such an
approach would jeopardize the Agency's effort to integrate data across all Agency systems through the
use of standardized data. Because we do not believe that the OIG is willfully endorsing a stovepipe
approach to data standards, we have provided revised language to correct this impression.
Our review also noted some minor factual inaccuracies for which we have provided corrected
language or comments in the attachment.
Attachment
26
Report No. 810017"
-------
APPENDIX IV
Page 1 of 1
DISTRIBUTION
Office of Inspector General
Acting Inspector General (2410)
Assistant Inspector General for Audit (2421)
Divisional Inspector General for Audit, Central Audit Division
Director, ADP Audits and Assistance Staff (2421)
OIG Water Program Coordinator (2421)
OIG, CAD Water Audit Theme Coordinator
Headquarters
Assistant Administrator for Water (4101)
Chief Information Officer (3101)
Assistant Administrator for Enforcement and Compliance Assurance (2201 A)
Associate Administrator for Reinvention (1801)
Associate Administrator for Communications. Education, and Public Affairs (1701)
Associate Administrator for Congressional and Intergovernmental Relations (1301)
Deputy Assistant Administrator for Water (4101)
Acting Director. Office of Information Resources Management (3401)
Director. Center for Environmental Information and Statistics, OPPE (2163)
Senior Information Resources Management Officer, Office of Water (4102)
Agency Audit Follow-up Official (2710)
Agency Audit Follow-up Coordinator (2724)
OW Audit Liaison (4102)
OIRM Audit Liaison (3401)
OARM Audit Liaison (3102)
Headquarters Library (3404)
27
Report Ho. SlOO!'
-------
------- |