1 Assessment of Non-Timber Functions: Information Needs
1.1 The far side of Designing Integrated Inventories: People and Politics H. Gyde Lund
Abstract
Multiple resource inventories in the USDA Forest Service are relatively new. People both as individuals and as part of an organization affect the design of an inventory. The politics of designing a successful inventory require an established vision, building an appropriate team, working together, establishing an information system, developing the data collection system, creating an appropriate administering unit, sharing information and securing funding and support.
Keywords: Inventory, multiple resources
1.1.1 Introduction
The theme of this conference is multiple resource inventories (MRI) especially the assessment of non-timber resources. The focus of our discussion is forest survey planning - should we prefer simple sampling designs to complex, but efficient sampling designs. There are other considerations that we have to deal with before we can settle on design. These include the people one needs to work with and the politics of actually getting an inventory underway - the subject of this paper.
This report is based on input from my colleagues in the USDA Forest Service (FS) and elsewhere and upon nearly twenty years of personal experience in trying to develop integrated inventories. Before we go on however, we need to define some terms.
An inventory is an accounting of goods or services on hand. We conduct inventories to present factual data on the state and condition of resources for assessments. From these data, analysts derive processes or changes in the resource base. We may be more interested in the processes that are going on, but we tend to sample for a state of the resource (PELKKI 1994).
An assessment is the act of officially estimating the value or character of property. It is the process of estimating or determining the significance, importance or value of something.
A data base is a collection of interrelated data, often with controlled redundancy, organized according to a scheme to serve one or more applications. The data are stored so they can often be used by different programs with little or no restructuring or reorgan
ization of the data. A data base management system takes a single underlying design for the data base and presents each user (or program) with a unique user schema (PELKKI 1994).
A successful data base is one that provides the principal users and stakeholders with the economic, social, and environmental information that they need to make sound and timely decisions and in formats they can understand and use.
Multiple resource inventories (MRI) are data collection efforts designed specifically to meet part or all of the information requirements for two or more functions or sectors (LUND 1986). They offer advantages over single functional inventories. MRI's are more economical and provide more comparable data across the inventory unit.
Integrated inventories - data collection efforts that are designed to link multi-sectors, data collectors, and decision levels over time (LUND 1986).
1.1.2 Background
Multiple resource inventories are not new. When human beings first evolved, they searched the landscape for areas that would provide food, shelter, water and fuel. They were, in fact, conducting multiple resource inventories. The goal was survival. As populations increased room to roam decreased. Humans had to settle down and start to devoting various pieces of lands to meet specific needs (e.g. agriculture, villages, timber).
Sectorial inventories developed focusing on the special uses of these lands. Now, however, our populations have increased to such a point that there are competing demands for the same terrain. We now need information for a multitude of potential uses.
Collection of data is costly and time consuming. Collecting the same information on the same piece of ground by different sectors exacerbates the situation. One way of reducing expenditures and getting complete and compatible data is to organize and join collection efforts through MRI. Thus, we have to return to the techniques of our ancestors. Today's goals of MRI are to promote communications between disciplines, improve data collection efficiency, and increase resulting accuracy {RECHTIN 1994) .
Forested lands in particular offer opportunities for multiple resource data collection.
For example, all forest lands have wildlife, economic values and ecological functions (air, water, and carbon sponges). Forest lands can serve as gene pools and media for maintaining or increasing biodiversity. Many forest lands are used for food, fuel, and fiber production and some forest lands are used for agro-forestry, grazing and recreation. Each of these potential uses offer opportunities for the development of MRI on forested lands.
MCCLURE et al. (1979), LU N D (1986), and RUDIS (1993b) have documented the concepts of MRI on forested lands. Most MRI have been conducted on forest and range lands in the United States, and most of these by the USDA Forest Service (FS). Some MRI work has been done abroad in Sudan (LUND et al. 1990, OBEID 1992), Tanzania (MGENI 1990), Australia (VANCLAY 1990), Senegal (GUEYE 1993), and Uganda (HEDBERG 1993 and DRICHI 1993). If an agency must manage its resources for more than one application and if temporal and scale needs are the same, MRI are useful in the following situations:
1. If base data are completely lacking.
2. If data exists but conflicts with other efforts or is an obvious waste of funds to continue individual efforts.
3. If some good information exists, but it is incomplete in "overlap" areas between sectors, consider a MRI to fill in. For example, forest inventories often do not include surveys of interspersed crop lands and vice versa. In order to manage these transition lands properly, complete information is really needed. Always build on existing systems.
Look for what is good and established and strive to make the existing systems more cost effective and utilitarian.
4. If agencies or ministries have established procedures and some ongoing efforts, but run into funding troubles, consider MRI to consolidate expertise, reduce field costs, and eliminates duplication of efforts. Again build on the systems that are sound and established. Coordinating inventories that fall under the jurisdiction of one ministry are easier to coordinate than those that fall under several ministries. In the case of the latter, each ministry must be made to feel a part of the MRI effort, benefit from the activity and share in the credit for its completion.
Impetus for MRI in the United States began in 1974 with the passage of the Forest and Rangeland Renewable Resources Planning Act (RPA). This Act directed the FS to periodically assess the forest and rangeland resource base of the United States in cooperation with other federal agencies. At the same time the law directs the federal agencies to coordinate and eliminate duplicate data collection efforts. RP A also directed the Secretary of Agriculture to prepare a renewable resource assessment not later than December 31, 1975. These data are to be updated every ten years. Since the passage of the Act, there has been greater emphasis on MRI in the natural resource area.
With the passage of the National Forest Management Act of 1976, the FS found that more site specific information and inventories incorporating mapping were needed for planning at the local level. Table 1 lists some inventories currently conducted on National Forest System lands within the FS.
We realized that no single inventory can answer all questions for an agency as large and diverse as the FS. It is neither possible nor necessary to develop a "One-Point in Time at the Same Place" field inventory to cover all resource needs. Some data may have to be collected on the same piece of ground by the same people but for different purposes. For example, some collection efforts are seasonal or cyclic in nature. A range specialist may conduct a vegetation inventory on a piece of terrain in the summer and a snow survey at the same location in the winter. It would be impossible to combine both surveys.
Some data need to be collected at specific locations such as water quality data at spring seeps, while other data need collecting throughout the landscape (e.g. soils, vegetation).
Some surveys, such as those of wildlife, may have narrow windows for collecting data or require staying in one place and observing animals over long periods. Many types of data collection require special skills that are in scarce supply or would be too costly to include on all inventory crews (RECHTI N 1994). Except for using common codes, definitions, and standards, these data collection efforts may be not be integratable with other inventories.
On the other hand, resource inventories that feed forest plans and national assessments, could be coordinated and in many situations, integrated in to a cohesive data collection strategy. For example, many sectors make use of existing vegetation data, such as forestry, wildlife, agriculture, range, and recreation (see table 1). These interest groups may collect similar information in the same areas. In many instances, these data collection efforts can be coordinated or integrated.
To conserve inventory funds, promote data sharing, eliminate information gaps, and to promote more consistency in resource mapping and inventory approaches, the FS issued direction to its field units to plan and conduct integrated inventories that feed the development of forest plans and national assessments (USDA Forest Service 1990). To assist in this process, standard definitions were developed for agency use (USDA Forest Service 1989). Our philosophy was to promote integrated and MRI yet provide maximum flexibility to the field. We assumed that if we outlined our agency-wide information needs and provided standards and definitions to meet those needs, then whatever methods the field chose to use would be acceptable.
Table 1. Resource inventories conducted on National Forest System lands (USDA Forest Service 1992).
Inventory Subject Major Use(s) Responsible Staff Vegetative
Component Nationwide forest National assessments, state survey Research - Forest Yes
survey reports inventory and analysis
(FIA)
Forest health National assessments, state survey Research - Forest fire & Yes
monitoring reports atmospheric sciences
Forestwide survey National assessments, state survey Timber management Yes reports, forest plans
Stand examinations Forest, project plans Timber management Yes
Timber cruises Project plans Timber management Yes
Regeneration surveys Project plans Timber management Yes
Research natural Project plans Research - Forest Yes
areas management
Range analysis National assessments, forest, project Range management Yes plans
Water quality National assessments, forest, project Watershed & air Yes
plans management
Soil resource National assessments, forest, project Watershed & air No
plans management
Air quality National assessments, forest, project Watershed & air Yes
plans management
Threatened & National assessments, forest, project Wildlife & fisheries Yes
endangered species plans management
survey
Wildlife & fish National assessments, forest, project Wildlife & fisheries Yes
habitat survey plans management
Cultural resources National assessments, forest, project Recreation management Yes plans
Recreation National assessments, forest, project Recreation management Yes opportunity spectrum plans
Visual management Forest, project plans Recreation management Yes Common variety National assessments, forest, project Minerals & geology No
minerals plans management
Fuels inventory Forest, project plans Fire & aviation Yes management
Forest pest condition Forest, project plans Forest pest management Yes
1.1.3 Challenges - The Far Side
We have found that our direction has not been enough. In a recent survey of our Regions and Stations, we found that most field units are collecting some information specified in the directives, but none are collecting it all. Of the standards and terms that are available, only two out of nine regions and two out of eight stations indicated that they were
following the direction fully. Some field units are not able to provide the data called for and some choose not to follow the definitions and standards provided for a variety of reasons (USDA Forest Service 1993).
If one compares resource data from one National Forest to another, we still find differences in the information available even where biological, physical and human situations are the similar. As a result, data between adjacent units are inconsistent and incomplete. Inefficiencies and duplication abound in both sampling design and data collection activities. Some inventories are not well focussed on answering critical questions and meeting critical modern objectives. Inventory data is being used inappropriately to answer questions it was not designed to address. Some key areas are being ignored, while we are collecting reams of information designed to answer questions that are either no longer relevant or at least are no longer high priorities. Much collected data are never analyzed.
In the same survey mentioned above, seven regions and four stations felt that current direction was not adequate and called for further instructions on how to design and implement coordinated inventories (USDA Forest Service 1993). Obviously, we are still having problems with developing integrated and multiple resources inventories.
1.1.4 People
Our field units identified three problem areas - these include the individual, the organization, and current inventory designs. All of these involve people's attitudes, perceptions, and willingness to work together.
1. Individuals. Some of the factors affecting the ability to effectively design MRI include:
- Many natural resource specialists are introverts. We chose our profession because we prefer to be alone. Consequently, natural resource specialists are independent and may have trouble working as team players (MADDEN 1993). MRI require team work.
Functionalism. Many specialists don't trust other disciplines. Environmentalists may be opposed to foresters collecting data on wilderness areas because they may assume foresters are looking at ways of converting the lands to timber productions. Industry and private groups may be afraid that data collected on rare and endangered species may lead to restrictions on the use of the land. Functionalism can also lead to a failure to consult other specialists about matters that the others may have dealt with for a long time. Failure to do so may result in reinventing the wheel by ignoring a lot of collective experience - whether it's functionally tainted or not.
In addition, some functional experts may declare knowledge or express demands on what and how to inventory but have no experience in doing the inventory work. This may lead to unrealistic expectations or results and distrust with those that bring up sampling difficulties or budget realities (LARSON 1994).
Tradition is the one of the biggest reason for insufficient coordination. Some methods and terms in use functionally by various groups within the FS have not changed since their start. There is general unwillingness by leaders and resource specialists to change definitions, standards, or procedures when it will disrupt the ability to analyze trends.
Perception of a change from one elevated resource to maintaining all resource areas on an equal level such as the change from a timber emphasis to watershed emphasis (MADDEN 1993).
- Fear of losing control/power over a project by letting other resource areas gather data (MADDEN 1993). Ownership in the process by all resource areas may be difficult to obtain.
- Use of unfamiliar terms and definitions. This leads to an inability: to share data, to tell other people what that information provides, and to clearly communicate the types of data they gather and how it is used in analyses.
- Ignorance, but not in a derogatory sense. Surprisingly, (or perhaps not so surprisingly) many people may be not fully aware of the direction and its advantages. Word does not filter down from the decision-makers to the field units.
- Lack of interdisciplinary or multiple resource inventory skills. Individuals are normally trained in one or two disciplines, not the several required of a MRI task ( R ECHTIN 1994).
- Lack of initiative or taking the lead. Direction is there, but field specialists are waiting for immediate supervisors to tell them to follow it. Also there is a reluctance to follow direction because it is viewed as dictated rather than being a more efficient way of getting information.
2. Organization. Field units identified the following problems relating to organization for MRI.
- A not-invented-here syndrome may occur within groups in an organization. Techniques or processes developed outside a particular group are ignored. Such a closed-shop approach fosters inbreeding and stymies innovation.
- Agencies, that historically have been single focused, have difficulty changing direction because of past successes and fears of future changes (SCHMIDT 1994). Associated with this is the fear of implications of issuing new direction. If decision makers issue new direction, such as consolidating inventory efforts, then the direction needs to be funded, implemented and enforced. If none of the above actions are taken, then the agency can be challenged in court for not following its own policies. On the other hand if the direction is implemented, one can challenge the direction, but not the data.
- Poor understanding and no consensus on the priority questions that need to be answered. These form the objectives for data bases, inventory, classification, mapping, and monitoring. Without clearly articulated goals, it is nearly impossible to develop appropriate sampling designs, etc.
- Poor coordination/communication between organizational levels and administrative units including differences in perceived priorities between the headquarters and the field units.
- Lack of support from within the agency for adequate time and money to do the job.
Focus on immediate outputs adds pressure to developing the best survey methodology.
There is a perceived lack of time to make changes from a single inventory to a multiresource inventory and a general feeling that it takes more time to think through a new process rather than just following the "old way" of doing business.
- Functional budgeting, attitudes, and approaches. There are often limited funds to do broad scale inventories other than for timber. Yet "timber dollars" may not be appropriate or appropriated to conduct vascular plant and soil inventories while collecting the overstory data. Funding and interest, by other disciplines, are low when the inventory is designed to survey timber production. Other agency interest and priorities result in lack of a true commitment in funding and completion of project.
These characteristics hamper appropriate resolution of the problems. They increase in severity in times of budget stress.
- The placement of inventory specialists in separate staff units, for inventorying natural resource basic data, presents real and perceived obstacles to integrated inventories, data collection, and information management.
- Lack of a strategic and coordinated inventory plan. Management does not recognize planning as something that needs to be done.
- Lack of enforcement or ability to enforce direction once a decision has been made.
3. Design. The following comments were noted regarding design of MRI.
- Most inventory efforts concentrate on vegetation structure and composition. The elements of system functions or processes for ecosystem management may not be present. The following subjects often need to be considered: biodiversity assessment from species levels (including soil micro-organisms and vascular understory vegetation) to landscape levels, remotely sensed reflectance values for landscape scale monitoring, and characterization of non -vegetated and non-forested ecosystems. A challenge is designing inventory systems that are dynamic so, as the understanding of social and biological components of ecosystems develops, we will have access to the information we need. This is an extremely difficult task. It indicates the importance of acquiring the technology that would allow us to "re-inventory" more regularly as part of our information strategy. This technology includes the use of remote sensing and geographic information systems (GIS).
Emphasis on using what is already working for one particular resource. Add-on-type inventories may not work for all resources. For example, timber has driven the inventory process for most of the FS. Timber surveys make use of aerial photography.
The photographs are selected to emphasize the timber resource. Color and tone on the photographs are selected for identification of tree species. The time of photo flights are tied to optimum contrast for tree identification. The scale of photos is established for tree identification (JAMES 1993). Other resources may have need for different kinds of imagery.
Many timber inventories make use of variable radius plots or point samples. This plot configuration may not adequately address the development of classification and the physical mapping processes that are frequently included under the title of "inventories"
from other resource viewpoints. This adds confusion to the issue. Objectives have changed, new partnerships need to be built and cultivated
Available timber information may not be not suited for a particular use such as for a goshawk habitat survey. The existing information can cause the specialist to go to the wrong places, reach the wrong conclusions, and waste time (RECHTI N 1993).
On the other hand, abandoning information or converting the existing information are costly enough to tempt agency managers to say no to the new design (LARSO N 1 993).
- Inappropirate scales - some sectors, such as wildlife, are used to working only at the local level while others, such as timber, have designed inventories to meet needs at the state, national and international levels. The problem of scale is one that always forms a major obstacle to integration (HANSEN 1993). Users of timber data, like the FS's Forest Inventory and Analysis (FIA) Units, have found ways to use a single inventory to solve multiple problems (answer questions) for a wide range (various scales) of geographic concern. Everything from global to national down to regional and state analyses have been done with FIA data. The claim of needing less detail at a larger scale of analysis has validity only if all relevant resource relationships have been accounted for at a smaller scale of analysis. When data become too weak at the more local levels, management inventories or specially design inventories must be used.
Other sectors have a difficult time integrating with FIA surveys because using a single inventory to address issues of very different scales has not been overcome. Wildlife inventories vary by species of interest and scale of the question. Recreation inventories are very site specific and are geared to the users more than the resource. One must address the issues of scale before integrating various inventories.
- Lack of spatial or geographic linkages. If resource characteristics cannot be linked to geographic location then it is not possible to assess the degree to which any one resource affects, or is affected by, other resources. Extremely important management costs linked to location cannot be considered. In short, resource data that are not loca
tion specific will not support a Forest plan which has any hope of complete implemen
tation.
- Lack of standard definitions and objectives. There may be both common and dissimilar attributes collected by each sector. Where methods are different, surveys may not be compatible and may not be able to be grouped together unless they have compatible goals. The distinction between classification development, and mapping and inventory - and the processes used to produce each - are not uniformly understood within and between all levels of the agency. These misunderstandings prohibit effective communi
cation and resolution development.
- Failure to test the system adequately before implementing.
1.1.5 Politics
"Politics" is the use of intrigue or strategy in obtaining a position of power or control. This section deals with recommendations on how to deal with the challenges mentioned above.
1. Establish a Vision. First, there must be a vision, mission, and objective where the agency wants to go with resource management and the information system that will support that vision (ELLEN 1993). Top level management should communicate its priority to those who will eventually design, use, maintain, and benefit by the integrated system. These are the ones who must understand the concept and make significant contributions to its development. Decision makers and line officers must get involved to the point where they know what is going on (ELLEN 1994). Recognition of the need and allocation of resources by line officers will be required before an attempt at coordination is possible. They must get the message that the task is important, that it has priority over everyday chores, and that the field is expected to contribute.
The objectives for the data base and the multi-resource inventory and data base must be determined by the decision-makers and resource managers. The goals may be to collect basic data for agriculture, forestry, livestock, wildlife, and watershed management and to establish a system for monitoring changes in the base for responses to treatments. A focus, such as ecosystem management, is needed that diffuses functional attitudes and approaches.
The process begins by clearly identifying the major land management questions needing answers ( establishing objectives). This could best be arranged by each level of the planning process and along an ecological hierarchy. The line officers should identify and prioritize the critical questions. The desired information is worked out at a conceptual level (LARSON 1993):
a. The need for a comprehensive, holistic data base to cover such areas or items as:
wilderness, old-growth, all vegetation (range, woodland, forest vegetation), water, and
soils. This may best be carried out using several individual "specialty" teams, working separately under a common integrated data analysis objective - knowing where and how the data they collect fits into a single data base, and how those data are to be integrated in the ecosystem analysis process.
b. The need for a formal, coordinated approach to bring data collection efforts together.
c. The need to continue to build uniform direction and support for local "holistic" data collections and integrated data analysis processes.
2. Build a Team. An interdisciplinary team is needed to translate the vision and objectives into an inventory program. Multiresource inventories require multiple approaches and input from other disciplines in an atmosphere of trust (RUDIS 1993a).
To be politically successful, "coordinated" is the key word. It is possible that the coordination would be simply the facilitation of getting the appropriate parties together.
Coordinated planning, design, and development of resource data bases, however, should occur. Involvement must be conducted in an interdisciplinary fashion. All interested user groups must have the opportunity to provide input into the process and have their needs heard and addressed (PELKKI 1993). The end-users (field foresters, wildlife biologists, resource specialists, etc.) must have ownership (involvement) in the inventory design process.
Get all the necessary people involved early. The task of the team is to establish common linkages such as standard definitions, methods for measuring the common elements, and units of measure. The team should also establish a common process geographically registering and storing the inventory data and a common process for managing the data.
Seek team members that are knowledgeable, have authority to make changes, and are willing to make the necessary changes to carry out their charges. Each function must be willing to give up "ownership" of its data and share the data management and collection duties with others (P ELKKI 1993). In addition, seek team members that can stay with the job until the task is completed. Educating new members can slow the progress. Where skills are lacking, provide training (RECHTI N 1994).
3. Work Together. "All of the troubles of man are caused by one thing, which is their inability to stay quietly in a room" - Blaise Pascal. Some suggestions to move teams to consensus include:
a. During team meetings, have a fixed agenda. Strive for decisions from each meeting and follow up actions. Document every decision and keep files on the resolutions. Circulate minutes and frequently invite comments from the field. This strengthens the program, keeps others informed and prevents surprises, builds support for the program, and alerts others who may be on parallel tracks of opportunities to coordinate. Document to whom the minutes or notes were sent and when, the comments received and follow-up actions taken.
b. Keep the objectives of the data base and inventory in mind throughout deliberations.
Keep the goals in focus. Then develop the design. Don't start with the design and make the objectives fit it. Construct the data collection system to meet the goals, even though it initially may not be the most efficient system that one may develop. Methodologies will have to evolve over time.
c. Keep an open mind. What one sector thinks may be an improper method of gathering data, may be fully acceptable to other disciplines and vice-versa. Design data collection methods using scientifically (as opposed to statistically) valid methods that are consistent with previously defined decision reliability. Recognize that the same sampling design may not work for all resources. For some, sampling may not work at all. Many disciplines may not have a feel for statistical sampling and may not be able to use it effectively. For example, soils surveys are a combination of science and art. That is acceptable to those in the soils profession. Team members need to be sensitive to the practices of others and incorporate those methods into the design.
d. No one resource group, or a cabal of several, should steamroller the rest and impose their favored techniques at the expense of others. Everyone must participate and be heard.
No "ecological snobbery" or impugning of motives should be allowed (B RICKELL 1993).
Speak to seek. Don't preach - listen. Try to understand what others have to say. Always provide constructive feedback when requested. Say what needs to be changed and provide the new wording so there is no misinterpretation.
e. Start simple and take small steps and implement in the same way.
f. Work from known to the unknown and move from what can be agreed upon to the more complex.
4. Establish the Information System. Build the information system first. A key to the issue is understanding the relationship between inventory system(s) and information structure(s). Regardless of the number of inventory systems used, the data must go into one information structure with one set of standards. Failure to require this means splintered and often duplicated information throughout all levels.
Of equal importance is the relationship between inventory for forest planning and agency needs, and inventory for project level. The information structure must be flexible enough to handle both forest planning and agency level inventories, as well as larger scale, project level information. The flexibility also must account for the variability that occurs throughout the agency.
a. Agree on the objectives. Define the decisions that need to be made using inventory data including the degree of reliability. Agree on the inputs to generate outputs. Concentrate efforts on developing an efficient, workable, single repository for data to be used in ecosystem analyses and focus on data elements that are not geographically limiting {SCHMIDT 1994).
b. Exchange information by each resource area on the type of data that they are now collecting and why. It is important to share how they use these data and the type of answers they provide.
c. Based on the above, develop a common data base and record keeping system. As an example, soils scientists need to collect soil survey data on ground cover so fire specialists also can use it to analyze fuel loading and wildlife biologists can use it for their analysis (MADDEN 1993). The data base system should present information so decision-makers and the public can readily understand. Include measures of the quality of data behind the analysis in the presentation of the results.
d. Agree on definitions of terms. This provides opportunities for all to be talking using the same terminology rather each resource's lexicon.
e. Agree on standards and acceptable errors. Where there is a large diversity within the range of the administered lands, the standards must be flexible enough to encompass this diversity. Chose the amount of acceptable error carefully because it has major implications for the appropriate sampling design, sampling methods, and costs of each. Arbitrary decisions on error levels could lead to bank-breaking costs or to the collection of useless data. Specific agency-wide standards for error may be appropriate for broad, agency-wide objectives but generally, direction should be limited to broad statements of intent and policy. The amount of error we should allow will vary depending upon the nature of the question being answered. The answer should depend upon risk (of a variety of types) and benefits/costs.
5. Develop the Data Collection System. Look at objectives, needs, time, funds and people.
Look at potential costs of various options, then agree on design. Compare what sectors are gathering data that may answer questions and needs of other resource areas. Look at where redundancies occur. Determine if one can collect information to meet the needs of mulitiple users and at the same time so that it does not slow the inventory process (RECHTIN 1994). This not only cuts redundancy of field surveys but provides a data base that is common for several resources.
a. Start simple and with a system that can be easily implemented and understood. Add to the system as needs and capabilities arise.
b. Incorporate available technology (Landsat/Thematic Mapper, aerial videography/image processing, etc.) into the data collection process. The integration of new and old techniques and technology in developing countries is innovative. Where we, in the developed countries, tend to use one or the other, they integrate and use both. Where we would use stand inventory data from field examination or use remote sensing data, they would integrate the data taking the best from each method and use that. They don't say "it won't work" but "let's try and use what will" (P LATT 1993). We need to take the same attitude in the rest of the world.
The use of remote sensing and GIS technologies provide techniques that would aid in classification, mapping, and inventories of ecosystems. These techniques allow stan
dardized approaches across large areas, increasing compatibility of procedures. Other levels of information could be added in a GIS format.
c. Test and revise system. Scrutinize proposed inventory efforts to determine (1) if the information is not already available and (2) if the required information is not too detailed in nature for the intended use. Explore opportunities to interpret, stratify, classify, and extrapolate existing information before instituting additional inventories. Every field inventory should receive interdisciplinary review before being conducted to maximize efficiency and avoid duplication. Field testing the survey methodology as data are gathered is important. Take your statistician, computer programmer, decision-maker, resource administrator and potential critics to the field with you on a demonstration plot. Refine the system needed. Build quality assurance and control into the process including training standards, consistency checks, and close supervision.
d. Develop a strategic plan for carrying out the inventory. The development of a strategic plan could have tremendous benefit. "Band -aid" approaches that make minor changes of the current framework will likely produce only very short-term, inefficient solutions.
6. Funding and Support. Major funding for inventory in the FS has traditionally been through timber, with small amounts for soil resource surveys. Establishing a generic inventory budget line with direction to create an ecological resource information base, would shift inventory into an integrated light. The development of a resource-combined budget for inventory, survey and monitoring will help reduce functionalism brought on by many current budgeting processes.
Depending upon the scale of the question being addressed, the agency may need the cooperative participation of adjacent and sister units, outside interests, universities, other agencies, other governments (States), and other interested parties and cooperators.
"Those who pay the freight gets what is delivered" (SCHMIDT 1993) and those who benefit from the information should pay (PELKKI 1994). It is usually industry and the timber interests who support resource inventory programs with money, in-kind assistance, or lobbying efforts (KI NGSLEY 1993). Often secondary data are gathered but the driving force behind the inventory is the source of funds. When funds are limited, one has to be selective about which data are measured and evaluated. To gain support (POWELL 1993):
a. Develop a message justifying the data base and inventory system. Focus on benefits to interested parties (win-win). Be persuasive and persistent.
b. Enlist cooperation. Identify key external interest groups and agencies as well as non-traditional groups and key internal leaders, individuals, and champions.
c. Seek recognition that all resources and land uses are consequential, that all resource programs are significant, and that the PEOPLE involved with the various resources are all important.
d. Keep everyone informed and involved - management, specialists, core teams, extended teams, users, maintainers and versed in the language and methodologies - before anything is started.
Funds, however, may not entirely be the issue - it is the need and desire to work together. As noted in the beginning of this paper, some of the poorest and least developed countries are actively using MRI. Such countries have no excess funds and have to look at inventory problems in a coordinated manner, determine what data are needed, access the available means to gather and analyze that data, select the most cost effective means, and then do it. Because they have frequently no entrenched bureaucracy defending an old established way of data collection and inventory, developing countries can quickly adopt and adapt new technology (PLATT 1993). If we can conduct the multitude of functional inventories that we now do, then we certainly have sufficient funds available to conduct MRI.
7. Create an Appropriate Administering Unit. In a large organization, the most efficient way to coordinate and integrate data and data collection is to centralize the efforts.
Restructuring an organization to include sector inventory specialists (for example, timber, fish, wildlife, soil, water, and ecology) under one staff group reduces functionalism. It also allows maximum cross-walking of data analysis needs, promotes consistency of timing and designs, optimizes "integrated" budget development opportunities, promotes overall consistency, and reduces duplication of efforts via closer day to day contact related to inventory and mapping programs.
There should also be a common control center and procedure for development and maintenance of definitions and standards within the agency. This center should have
responsibility for coordination outside the organization. Direction must come from a neutral, but knowledgeable source. Leadership by traditionally functional organizations and individuals defeats the purpose of integrated standards. A separate data management authority, that is not beneath any one particular group or sector, may be necessary. This group will be more objective and not subject to the whims or desires of any one department or user group. The data administrator should report on the same level as the line managers, giving the data resource the organization-wide support and authority necessary (PELKKI 1993).
Clearly assign responsibility for data administration, including proposed changes in data structures, definitions, and codes. Encourage the use of a corporate information system and discourage non-sharable data collection and storage. Data bases should contain the basic measured data to make interpretations.
Establish a well-defined review process for new terms and change of old terms. A well-defined process for submitting comments and changes regarding standardized terms and definitions is required. The process must clearly be one of facilitation, rather than dictation. It must follow a prescribed process to solicit input from all potential users of the data. During changes or the development of new data bases, both models must be kept operational (P ELKKI 1994).
Keep in mind, however, that in centralization, some specialists and staffs may view their jobs as threatened. Management needs to address these concerns. In addition, central
ization may decrease innovation in data collection technology.
8. Share Information. In the traditional organization, we often leave the interpretation of data to expert people. Most often the data are presented as objective, which they are not, and complete, which is an impossibility (WHEATL EY 1993, RUDIS 1993a). As we construct and travel down the information highway, we need: a broad distribution of information, viewpoints, and interpretations, organizational designs that foster multiple interpretations of the data, and systems that do not restrict information access. We should present solutions that transcend current organizational structures (RUDIS 1993a).
Integration needs to go beyond the survey stage and incorporated into the analysis stage (MADDEN 1993).
a. Present information in a form that is easily understood including the use of graphs, charts, computer "maps," and simulation and visualization techniques. In addition, close the information loop. Present the data in a form so that the data collectors can comprehended their importance. This is a final ratification and tells the inventory team that the data they collect have meaning (ROBI NSON 1994).
b. Provide unpublished raw data to anyone that requests it (RUDIS 1993a).
c. Encourage diversity in resource analysts through additional training and recruitment in non-timber specialties. Involve the public and especially special interest groups in the analysis of the data. Such groups generally have the skills needed to do an adequate job and by having them involved at the outset could avoid some surprises later on (RUDIS 1993a).
d. Use common work stations so the people who are gathering data and the people using the resulting information are in the same area. Use a common work room to promote team building and so people can easily discuss how the integrated resource inventory is used.
e. Fund worthy research proposals that make use of the inventory data and sample design through co-op arrangements (RUDIS 1993a).
f. Encourage multi-disciplinary interaction in resource publications, survey plans, sampling designs, email postings, etc. (RUDIS 1993a).
g. Formally keep track of what uses are made of the inventory data in nontraditional disciplines - an essential part of any pioneering research activity such as that by RUDIS (1991, 1993b ).
1.1.6 Conclusions
No single inventory will answer all questions in an agency as large and diverse as the USDA Forest Service. Opportunities for integration are the use of common definitions, noting and storing location of where data are collected, by whom, when, and for what purpose and objective sampling methods. Surveys which collect data on existing vegetation offer particular opportunities for integration (see table 1).
Obstacles to acheiving integration include individuals, organizations, and existing designs.
Success may be achieved by working with diverse people and groups, establishing a vision, establishing an information system, developing and testing a data collection system, seeking funding and support, creating an appropriate organization, and sharing resulting information.
Team work and commitment by all levels in the organization are key to the success of a multiple resource inventory. Having knowledge of the resources, however, is just one step in the process for successful resource management. Plans for land use have to be worked out in concert among the various sectors. Holistic assessments followed by integrated and coordinated planning and implementation is the only hope for determining the optimum use of Earth's limited resources.
1.1.7 Acknowledgements
My thanks to my fellow FS employees - Pete Avers, Ann Bartuska, Diane H. Burbank, Joan Comanor, Chuck Dull, Pat Durst, Brian Geils, Jerold Hahn, Lew Jump, Leon Liegel, Gary Man, Roy Mead, Teri Raml, Fred Samson, Mike Srago, Phil Weber, and John Zerbe - for their very useful comments. Special thanks are due to my friends from outside the Forest Service - Jeff Burley, John Kershaw, and Michael Kohl - for their helpful suggestions.
1.1.8 References
BRICKELL, J., 1993: Personal correspondence. USDA Forest Service, Northern Region.
DRICHI, P., 1993: The inception of monitoring woody biomass resources in Uganda-Jinja urban area as a case study. Paper presented at the UNEP and IUFRO International Workshop in cooperation with FAQ on Developing Large Environmental Data Bases for Sustainable Development. 14-16 July 1993. Nairobi, Kenya. (Proceedings in press).
ELLEN, D., 1993, 1994: Personal correspondence. USDA Forest Service, Alaska Region.
GUEYE, S. , 1993: Concise description of the Senegal forest inventory situation. Abstract. In:
NYYSSONEN, A., ( ed.): Proceedings of FAO/ECE meeting of experts on global forest resources assessment in cooperation with UNEP and with the support of FINNIDA (Kotka II); 3-7 May 1993; Kotka, Finland. Res. Paper 469. Helsinki: Finnish Forest Research Institute. 200 pp.
HANSEN, M. , 1993: Personal correspondence. USDA Forest Service, North Central Forest Inventory and Analysis Unit.
HEDBERG, C. , 1993: Development of a large scale forest national database of Uganda. Paper presented at the UNEP and IUFRO International Workshop in cooperation with FAQ on Developing Large Environmental Data Bases for Sustainable Development. 14-16 July 1993.
Nairobi, Kenya. (Proceedings in press).
JAMES, A. , 1993: Personal correspondence. USDA Forest Service, Mendocino National Forest.
KINGSLEY, N., 1993: Personal correspondence. USDA Forest Service, North Central Forest Inventory and Analysis Unit.
LARSON, M., 1993, 1994: Personal correspondence. USDA Forest Service, Southwest Region.
LIEGEL, L.H., 1993: Personal correspondence. USDA Forest Service, Pacific Northwest Research Station.
LUND, H.G., 1986: A primer on integrating resource inventories. Gen. Tech. Rept. W0-49. U.S.
Department of Agriculture, Forest Service. 64 pp.
LUND, H.G.; MOHIL E L D EEN, F.A.; ALLISON, R.P.; J ASUMBACK, T., 1990: The Sudan reforestation and anti-desertification (SRAAD) project: applications for watershed management planning. Paper presented at the U.S. Agency for International Development/Bureau for Science and Technology Workshop on Environmental and Institutional Assessments for Watershed Management Planning. 26-27 June 1990. Washington, DC. 11 pp.
MADDEN, Carolyn , 1993: Personal correspondence. USDA Forest Service, Stanislaus National Forest.
MCCLURE, J.P.; COST, N.D.; KNIGHT, H.A., 1979: Multiresource inventories: a new concept for forest survey. U.S. Department of Agriculture; Forest Service. Research Paper SE-191. Asheville, NC: Southeastern Forest Experiment Station. 68 pp.
MGENI, A.S., 1990: Forest resources assessments in Tanzania: current inventory and monitoring methods applied, problems and possible futurology. In: LUND, H.G.; P RETO, G . , (eds.):
Proceedings Global natural resource monitoring and assessments: preparing for the 21st century.
24-30 September 1989. Venice, Italy. Bethesda, MD: American Society for Photogrammetry and Remote Sensing: 546-556.
OBEID , S.M.H.; HASSAN, A.A.E., 1992 : Sudan resource assessment and development (SRAAD) first interim report. Khartoum, Sudan; Forest National Corporation. 36 pp. w/maps.
ORLAND, B.; DANIEL, T.C.; LYNCH, A.M.; HOLSTEN, T., 1992: Data-driven Visual Simulation of Alternative Futures for Forested Landscapes. In: WOOD, G.; T URNER, B., ( eds.): Integrating Forest Information Over Space and Time - Proceedings of the IUFRO Conference; 13-17 January 1992, Canberra, Australia. Anutech Pty. Ltd. 368-378.
PELKKI, M., 1993, 1994: Personal correspondence. University of Kentucky.
PLATT, B., 1993: Personal correspondence. USDA Forest Service, Mark Twain National Forest.
POWELL, D., 1993: Personal correspondence. USDA Forest Service, Forest Inventory, Economics, and Recreation Research Staff.
RECHTIN, Julie, 1993, 1994: Personal correspondence. USDA Forest Service, Modoc National Forest.
ROBINSON, A., 1994: Personal correspondence. Australian National University.
RUDIS, V.A., 1991 : Wildlife habitat, range, recreation, hydrology, and related research using forest inventory and analysis surveys: a 12-year compendium. Gen. Tech. Report SO-84. New Orleans, LA: U.S. Department of Agriculture, Forest Service, Southern Forest Experiment Station. 61 pp.
RUDIS, V.A., 1993a: Personal correspondence. USDA Forest Service, Southern Forest Inventory and Analysis Unit.
RUDIS, V.A., 1993b: The multiple resource inventory decision-making process. In: LUND, H. G.;
LANDIS, E.; ATTERBURY, T., (eds.): Stand inventory technologies 92 - Proceedings. 13-17 September 1992; Portland, OR. Bethesda, MD: American Society for Photogrammetry and Remote Sensing: 180-192.
SCHMIDT, T., 1993, 1994: Personal correspondence. USDA Forest Service, North Central Forest Inventory and Analysis Unit.
USDA Forest Service. 1989. Interim resource inventory glossary. Washington, DC: U.S. Department of Agriculture; Forest Service. 96 pp.
USDA Forest Service, 1990: Resource inventory handbook. Zero code, Chapter 10, Chapter 20. FSH 1909.14. Washington, DC: U.S. Department of Agriculture; Forest Service. Misc. pagination.
USDA Forest Service, 1992: Forest Service resource inventories: an overview. Washington, DC: U.S.
Department of Agriculture; Forest Service; Forest Inventory, Economics and Recreation Research. 39 pp.
USDA Forest Service, 1993: Issue Paper 3 - Uniform processes for ecological classification and resource mapping and inventories. In: LUND, H. G . {ed.) Proceedings National Workshop - Integrated ecological and resource inventories. 12-16 April 1993; Phoenix, AZ. WO-WSA-4.
Washington, DC: U.S. Dept. of Agric.; Forest Service; Watershed and Air Management: 68-79.
WHEATLEY, Margaret J. , 1993: A quantum vision: chaotic organization must replace the Newtonian bureaucracy. Eco-Watch. 12/1/93.
V ANCLA Y, J. K., 1990: Integrated resource monitoring and assessment: an Australian perspective of current trends and future needs. In: LUND , H.G.; PRETO, G., (eds.) Proceedings - Global natural resource monitoring and assessments: preparing for the 21st century. 24-30 September 1989.
Venice, Italy. Bethesda, MD: American Society for Photogrammetry and Remote Sens.: 650-658.
