Scope 47- Long-term Ecological Research, An International Perspective

14.1 BACKGROUND

The first Berchtesgaden workshop brought together scientists from various countries. Through the workshop, scientists were able to identify common ecological processes being studied at two or more sites, and to recognize that certain broad climatic phenomena were exerting similar influences on different ecological habitats. The scientists also realized the need for much broader geographical comparisons and analyses if ecological principles were to be identified and tested.

Workshop II built on the results of the Berchtesgaden workshop and on the personal and professional contacts that were fostered there. The goals of Workshop II were (1) to begin the formal development of international co-ordination and networking of collaborative long-term ecological research at specific sites, (2) to facilitate the introduction of the newest technologies available into these collaborative research efforts, (3) to identify initial working groups in three selected biomes, and (4) to develop specific plans for the international collaboration necessary for long-term ecological research throughout each of these three biomes. Long-term research questions identified during the first workshop in the Federal Republic of Germany provided a starting point for Workshop II. The research plans developed from the biomes included the long-term ecological phenomena to be studied, the comparable data to be collected, and the analytical techniques to be used in designing the studies, collecting the data, and sharing the information.

A conclusion that was drawn from both workshops was that long-term ecological research on the international scale depends upon the dedication of individual scientists, but that enhanced communication of ideas by the scientists is vital to the research. This chapter presents a narrative transcription of the proceedings of Workshop II, preserving the identity of the participants and their ideas.

14.2 PURPOSE OF THE SECOND INTERNATIONAL  WORKSHOP

The Albuquerque workshop (Workshop II) was considered a 'working' workshop; the number of participants was small, and each had very specific responsibilities. There was an implicit assumption that participation in the workshop mandated subsequent collaborative efforts to secure funding for and participation in long-term ecological research programs in three biomes identified at the Berchtesgaden workshop as being particularly amenable to international collaborative investigations. The three chosen biomes are (1) arid/semi-arid, (2) temperate forests, and (3) boreal forest/tundra. During the International Biological Program (IBP), some efforts were made toward international collaboration in these three biomes, but the efforts were largely post-facto comparisons of data which were not originally collected for international comparative purposes from projects which were not conceived in a collaborative fashion. In addition, at the time of the IBP, many of the current measurement, analytical, and communication technologies were not available; our concepts of global phenomena were more primitive, especially as they relate to ecological processes; and the use of models was not well developed. Subsequently, we have recognized the need to plan international research rather than just compare data from projects which were designed for another reason in different areas of the world. We have also learned that models are particularly effective in building collaborative research projects among scientists from various countries and ecological conditions. Thus, the development of specific, collaborative, international research by participants of the Albuquerque Workshop was designed to go beyond IBP, using research questions posed during the first workshop at Berchtesgaden, more sophisticated technologies, and greater understanding of the development and use of ecological models, in an additional decade or more of ecological research in the three biomes chosen for consideration.

14.3 WORKSHOP STRUCTURE

Before the Albuquerque workshop formally began, an all-day field trip was made to the Sevilleta Long-Term Ecological Research Site south of Albuquerque. The purpose of this field trip was to stimulate thoughts about collaborative long-term research among sites while actually being on a site, to demonstrate sophisticated methods of making large-scale field measurements, such as the Fourier Transform Infrared Spectrometer (FTIR), and to stimulate communication among the conference participants. The Sevilleta LTER site is one of the newest in the LTER network sponsored by the United States National Science Foundation. This site is unique in its location at the junction of at least four major biomes, a location which allows quantification of (1) gradient relationships with distance, (2) the scale-dependent or independent nature of spatial variability, (3) the influence of steep environmental gradients upon system properties, and (4) integrated responses across the region. Since research projects and the quantification of parameters at the Sevilleta occur at many different scales, what is learned about the studied phenomena, data management, and project synthesis should be helpful in the development of models for connecting local, regional, and global research efforts.

The first and second days of the Albuquerque workshop included presentations from discussions of the ecological questions that are amenable to long-term research on each of the three biomes. The presentations were not simply descriptions of current research, but rather made an attempt to identify ecological questions or hypotheses that must be addressed in an international and long-term context. These were followed by presentations of technologies that are integral to international long-term ecological research. The Great Plains modeling experience was presented by William Parton; the integration of small to large data bases to model regional and global CO₂ levels and predicted changes was presented by Michael Farrell of the Oak Ridge National Laboratory; geographical information systems (GIS) coupled with a dynamic modeling program (ARC/DYNAMO) were developed by Wolfgang Grossman and Jorg Schaller of Environmental Systems Research Institute (ESRI); remote-sensing, data management, and data networking, such as the NASA First Integrated Field Experiment (FIFE) project, at the Konza prairie LTER in Kansas was presented by Donald Strebel of Versar, Inc.; and new communications technologies selected to address a series of spatial and temporal scales were presented by Steven Storch of BBN Systems and Technologies Corporation.

Three of the technologies are appropriate for multi-scale research: the GIS-model technology for studying site-specific processes and integration at the site level; remote sensing and management of data from several integrated technologies to expand from the site to the regional or the continental scale; and communications technologies for sharing the results from the first two technologies to ensure the active continuation of research projects. This plenary segment of the workshop acquainted participants with research efforts in each biome and with innovative technologies to facilitate data collection, management, sharing, and real-time and computer-to-computer communication among scientists throughout the global community.

Three working groups composed of scientists who were intimately involved in long-term ecological research and who have access to or control of a substantial dedicated research site were then established. These working groups, which met at the end of the first and second days of the workshop, were given five tasks. They were asked to:

1. Develop one to a few research questions or hypotheses identified by the Berchtesgaden group that are international in scope, are also regionally important, and require long-term ecological research for a resolution;
2. Plan an experimental approach which ensures that international comparisons and analyses are used to their greatest power;
3. Describe technologies to be employed for both investigation and project communication purposes;
4. Provide a plan for the analyses of the resulting data which co-ordinates structural components for the proposed study of the three sites, allowing for the later addition of new sites and investigators; and
5. Develop strategies to obtain funding for the anticipated long-term ecological research project.

14.4 COMMUNICATIONS FROM THE WORKSHOP

The final segment of the workshop consisted of a televised presentation of research plans by each of the three working groups to an audience of scientists watching at most of the 17 LTER sites in the United States and at the Oak Ridge National Laboratory in Oak Ridge, Tennessee. Following the first hour's broadcast, scientists at these sites were encouraged to send comments and questions via electronic mail (E-Mail), FAX, or telephone. During the second televised hour, workshop participants addressed these comments and questions.

Workshops are a popular and common way for scientists to interact and advance the science of their discipline. Unfortunately, since workshops work best if discussions involve only small groups of individuals, a cost-effective and successful workshop can accommodate only a relatively small portion of the scientific community. Subsequent dissemination of information usually occurs as reports, books such as this one, or word of mouth. These forms of communication are often significantly delayed and can suffer from inaccurate interpretation. Small workshops also fail to involve the broad scientific community in the creative first steps of developing research plans. However, cost-effective technologies, such as the telecommunications already in use by large international companies, can allow near real-time interactions with a larger segment of the research community. This component was added to the Albuquerque workshop both to demonstrate its usefulness and to communicate and receive feedback from other scientists. This input from many scientists at a wide range of sites by telecommunications was a secondary goal of the workshop. The television programs were duplicated for the additional audience of colleagues of the international ecological research community who could not be reached by the satellite down-link.

14.4.1 THE FIRST HOUR OF THE TELEVISION PRESENTATION

On the third day of the workshop, participants went on the air via satellite to the US LTER network and Oak Ridge National Laboratory to present the results of the discussions of three working groups concerning long-term research in the three biomes of arid/semi-arid, temperate forest, and boreal forest/tundra. Paul Risser began the program by stressing that scientists watching the televised program, and the studio audience, were all part of an experiment in communications among scientists who were at large distances on the planet from each other, but who were all concerned with and working on global-scale biological problems. Whereas 30 scientists attended the first Berchtesgaden workshop in 1988, and about 20 attended the Albuquerque workshop, participation by those in the scientific community was increased by the television link to over 100 individuals.

Jim Gosz described the presentations and discussion which had led up to the television presentation, and emphasized that since the presentation was a progress report only, and no final decisions had been made, input from the audience was extremely important. It was announced that there would be three reports, one from each of the working groups discussing research in the arid/semi-arid, temperate forest, and tundra/boreal forest biomes, followed by a discussion of research which could be performed across biomes.

14.4.1.1 The temperate forest biome

The temperate forests working group was chaired by Dr O. W .Heal (Scotland), and included Peter Beets (New Zealand), and Wolf Grossman (Austria). This group identified three interrelated issues (climate change, pollution, and management) for the temperate forest biome, and focused on climate change as a point of entry for research. The three sets of high priority questions were defined as follows.

First, what is the contribution of temperate forests to the global carbon budget? Can management of these forests help to control global carbon cycling? Research on these questions occurs at the site level where decisions are made about experiments on processes, parameters to measure, and how to validate results. Results from site-specific studies are aggregated to a regional information base and combined with other site-specific information in that region for global synthesis of information from other regions. Geographic information systems (GIS) and remote sensing techniques can be used to combine information from sites and regions. Long-term ecological research is appropriate because there is between-year variation, variation over longer natural or managed forest cycles (rotation), and variation between cycles in a biome consisting of long-lived dominant plant species.

Second, what is the contribution of forests to the global flux of the radiatively active gases such as methane (CH₄), nitrous oxide (N₂O), carbon dioxide (CO₂), the hydrocarbons, and ozone (O₃)? What are the feedback relations to the forest and to climate change, and can these relationships be quantified? Can management control or modify these interactions? This is a key topic because of the complex pollutant-changing climate interaction and because there is a range of management options for controlling the sources and sinks of the radiatively active gases (RAG). A simple physical study of sources and sinks is inadequate because biological processes are controlling these physical processes. A pollution-climate gradient can be used to examine environmental conditions that have existed or do exist as a result of pollution for exploring the variation in responses to climate change.

Third, what is the response of temperate forests to the change in CO₂ levels, temperature, and moisture resulting from climate change? How do the responses vary with different soil, nutrient, and moisture variability? Answering these questions will require manipulation of systems to examine the response of components of the systems, including the responses of individual species or groups of species. The temperate forest group concluded that a collaborative, cohesive world-wide program should focus on a few specific topics (for example, quantifying the sources and sinks of radiatively active gases and targeting a few selected responses of the forest). Since these fluxes will be different in each type of system, and will depend upon feedbacks to and from the changing climate, a network will be necessary to understand the global system.

14.4.1.2 The arid/semi-arid biome

The arid/semi-arid group was chaired by Uriel Safriel (Israel), and included Mary Seely (Namibia), Francisco Garcia-Novo (Spain), Carlos Montana (Mexico), Bill Lauenroth (Central Plains LTER, Colorado, USA), and Walt Whitford (Jornada LTER, New Mexico, USA). For biomes whose structure and function are driven by seasonal and multiple-year droughts, the central questions are (1) how does arid system stability interact with human impact to produce desertification? and (2) how will climate change influence the phenomenon of desertification? Finding answers to these questions depends upon understanding why human-impacted arid systems transform to non-exploitable, resilient states. What determines the resilience of impacted systems? 

The general hypothesis is that the interaction between the stability properties of the system (resistance to change) and the type and strength of human impact (force applied to the components of the system) determine the transition of an arid system from one state to another. Theoretically, if global climate change is ignored, human impact on the world's arid systems (grasslands, shrublands, and scrublands) should produce divergence in the original system (the product of such influences as precipitation quantity and timing and temperature) to different secondary states (dependent upon the force of human impact on the same ecotype in different places on the globe). As human impact on the new system continues or changes, there should be a convergence of properties in severely affected arid lands. If human impact is like the force exerted on a ball to move it out of a pocket representing system stability, then the new impact of climate change will primarily affect system stability represented by the shape and depth of the pocket. Arid climates are believed to be much more variable than more mesic ones, but arid ecological systems are more stable because individual species have evolved adaptations to the climatic variability. The influence of climatic change upon arid system variability and the response of arid species to changed variability and a more rapid rate of change is entirely unknown. It is also important to identify those systems which are candidates for transformation and those which are already transforming. The model of transition from one state to another recognizes that system structure and function may transform independently.

To test this hypothesis, both world-wide experiments and long-term monitoring are necessary, and international co-operation is needed to identify as many states as possible, and to evaluate the relative contributions of climate change and human impact on the stability properties of states and the transformations from one state to another. The experimental manipulations of these arid systems are climate modifications of the amount and timing of water availability. These can be accomplished by using irrigation, rain-out shelters, redistribution of runoff, and the simulation of human impact by removal or addition of soil and/or vegetation. Three important physical variables to monitor are soil organic matter, nitrogen mineralization, and water infiltration. Measuring organismal responses might include changes in plant life forms; individual plant and community architecture (important to the animal community); plant primary production, growth, and phenology; physiological responses; and biodiversity represented by species composition and richness.

Long-term research is important to determine the effect of unusual, unpredictable natural history phenomena and environmental events which are of distinct and overwhelming importance to arid lands. An international effort is vital to allow major existing states of arid land systems to be monitored simultaneously in order to understand the impact of global change as it interacts with human impact and systems stabilities world wide. World-wide research will permit monitoring of predicted changes in albedo, dust generation, and the activity of decomposer communities (important contributors to radiatively active gas generation).

14.4.1.3 The boreal forest/tundra biome

Gaus Shaver (Tundra LTER, Alaska, USA) chaired this group, which included Kari Laine (Finland), Don Strebel (NASA, USA), Mike Farrell (Oak Ridge National Laboratory , USA), and John Yarie (Bonanza Creek, LTER, Alaska, USA). Four major long-term research issues were identified:

1. The soil carbon budget and the quantity of carbon released as either CO₂ or CH₄ is virtually unknown in this biome, causing uncertainty about whether it is a source or a sink for carbon;
2. Boreal forest/tundra biomes are strongly decomposition limited, and the constraints on decomposition through nitrogen and phosphorus cycling may be very important in regulating carbon cycling processes and net exchange of carbon with the atmosphere;
3. Episodic, large-scale, and/or long-term phenomena such as animal population fluctuations and fire are important forcing functions for ecosystem dynamics; and
4. Permafrost and soil hydrology are keys to system stability.

Research needs in this biome are:

1. A network of monitoring stations based upon such guidelines as those just being instituted in the Nordic countries and presented in a publication called Guidelines for Monitoring in Nordic Countries by Kari Laine (Finland). This publication describes the standard protocols and close communications being set up in the Nordic countries to monitor biological change. The monitoring program was in part the result of the Nordic countries' lack of preparedness in the face of the Chernobyl catastrophe, which had disastrous effects on northern biological communities and organisms;
2. Identification of common experiments in which the same manipulation is performed across the boreal/tundra region to determine whether limiting factors in this area are the same; and
3. A well-developed boreal/tundra geographic information system which, combined with remote sensing, can monitor long-term change in response to fire or animal cycles. A network of monitoring stations, common experiments, and the GIS can all be used to model changes in response to global warming and new climate variations, and to validate and verify the model.

14.4.1.4 Cross-cutting research

Peter Beets (New Zealand) examined themes that were common to the recommendations of all three working groups. There were three of these. Biological responses to global climate change were recognized as species' responses, making it necessary, therefore, to understand how species' physiological responses to climate change are integrated from the community through the ecosystem level. The second theme was the role of biological systems in bringing about climate change, particularly by controlling the majority of atmospheric gases and by affecting water availability and distribution. Experiments in primary productivity and nutrient cycling were also themes common to all biomes. Last was the question of human impact on the biology of the globe, and the human responsibility to understand and to manage these impacts. Both scientists and policy makers must act together to become responsible for human impact and its implications to biological systems. Subsequent discussions noted that biodiversity and the frequency and intensity of disturbance were also common topics.

14.4.1.5 Reactions to the reports

Bill Lauenroth and Jerry Melillo expressed the belief that the importance of modeling in world-wide research efforts could not be emphasized enough. Lauenroth pointed out that global-scale research is now enhanced by improvements in technology such as remote sensing and computer power. Melillo emphasized that although links between GIS and dynamic models are difficult at present, they are possible. He noted that new techniques for research, monitoring, modeling, and validation must be put very clearly before professional scientists by those adept at these technologies, so that scientists can creatively conceive the uses for these technologies. Wolf Grossman encouraged the scientific community to be bolder in asking for immediate support for long-term global-scale research, pointing out that even if all polluting emissions could be stopped now, the situation world wide would improve only very slowly. Grossman expressed the belief that management at the ecosystem level was the only means of rapidly improving the situation once emissions were controlled, and this urgency should lead to more aggressive requests for research support. Walt Whitford made a plea for imaginative and large-scale experiments which, although expensive, can provide insight achievable only through whole-system manipulations. Don Strebel was asked whether NASA has produced enough information through remote sensing to use in making models. Strebel replied that there is much that NASA is doing in terms of modeling soils, vegetation, atmospheric behavior, and the links between these. Strebel emphasized that model verification and validation is needed. This requires intensive fieldwork to check links between components of systems over the short term, and to check the accuracy of predictions over the long term. Francisco Garcia-Novo also stressed validation, but suggested that historical and paleohistoric information be used to validate and verify model predictions. In such historical information, it is possible to trace past vegetation responses to climate change and human impact and management in order to predict landscape changes. Climate has changed in the past, and there is information in tree rings and in fossilized pollen to help us understand current changes.

Mary Seely pointed out that since there is a diversity of habitats and differing degrees of impact felt or predicted as a result of climate change, the few large and expensive experiments should be augmented by an extensive range of experiments that were site-specific. John Yarie focused on the problem of not having basic information in the boreal/tundra biome with which to build a predictive model.

The idea of large, whole-system manipulations or flagship experiments performed in all three biomes was important to Jerry Melillo because it allowed for comparison and contrast of a few common manipulations. Kari Laine said that it is sometimes difficult to predict which monitoring variables are going to turn out to be important in the future, and Bill Heal urged boldness in mounting large experiments in several biomes. In Europe, scientists should build on the Nordic Monitoring system and on successful German modelling of effects of air pollution on forests. In concluding the first hour, Paul Risser emphasized the significance of this time in the history of the world - a time when human impact has accelerated climate change through such things as land use - and pointed out that the new research is not just a question of understanding individual systems but is more a question of keeping the globe working. The central problem of arid systems is a lack of understanding of transitions from one state to another, in the temperate forests it is trace gases, and in the boreal/tundra regions it is soil processes.

14.4.2 THE SECOND HOUR OF THE TELEVISION PRESENTATION

14.4.2.1 Comments from viewers

Bruce Milne (Sevilleta LTER, University of New Mexico) reminded the panel that it is likely that plant communities may not respond as a unit to climate change, and that, according to paleoecological data, extant communities found in different biomes are recent assemblages of individual species. Lauenroth agreed that the response of individual species, rather than the entire plant community, should be a focal point of research, but thought that although extant plant communities may be quite transitory assemblages of species, responses to global climate change will first be seen in changed dynamics within an existing system structure. He also expects dynamics to change substantially, whereas community structure may not change for a very long time. The Central Plains Research LTER found the arid zone working group's conceptual model of arid ecosystems moving from one stability state to another to be unrealistic. They argued that ecosystems, especially arid ones, are not stable, and therefore the arid/semi-arid group's conception of an ecosystem transforming from one set of stability conditions to another was incorrect. Uriel Safriel stated that arid systems always have some properties of structure and function which are measurable and are known to change. Therefore, the question being asked in the conceptual model is, what is the kind of stability seen at present? It might be that a present system is at an extreme position of stability or lack of stability, but the important issue is how systems move under human impact from the present perceived state to another state. Milne suggested a refinement of the ball and pocket model presented by the arid working group. The ball is a golf ball which is moved in and out of a soft clay pocket. While it is in the pocket, however, the golf ball leaves small impressions in the soft clay which represent a different scale of stabilities within the pocket. Differences in variations in the environment which cause the ball to vibrate between different smaller states (the little depressions) within the pocket are not the variations in the environment responsible for moving the ball into or out of the pocket. Safriel agreed that the issue of scale brought up by Milne is predominant in all areas of ecology.

Jim Ellis (Central Plains LTER, Colorado State University) expressed the belief that the choice of desertification as the leading problem of arid/semi-arid systems is an overstated or highly questionable focus, whereas evaluating stability or shifts in system state resulting from human or climatic influence is quite reasonable. Francisco Garcia-Novo agreed that the overstatement of 'desertification' problems by some countries for political reasons was absolutely true. The word was chosen because it encompasses the acute problems that occur in arid and semi-arid areas. Usually 'desertification' applies only to the encroachment of arid areas by the desert, but this word was used by the working group to cover much broader problems in tropical, subtropical, Mediterranean, and arid areas that experience a dry period. These areas are suffering acutely from human impact. Where this impact has been very strong, it has produced very different results in different systems. A tremendous gradient of degrees of impact and effects exists from central Europe to central Africa, and obviously all impacts cannot be described as desertification.

Diane Marshall (Sevilleta LTER, University of New Mexico) suggested that the potential exists for the evolutionary response of individual species to alter and be altered by global climatic change, and this phenomenon deserves attention in the organization of long-term studies. Peter Beets answered that the temperate zone working group did touch on evolutionary studies when it discussed the physiological responses of individual organisms, but he agreed that evolutionary studies should be incorporated into LTER sites.

James Halfpenny (Niwot Ridge LTER, University of Colorado) pointed out that there is a difference in sensitivity to environmental change among various ecosystems, and suggested that alpine and arctic tundra systems are probably the most sensitive. Therefore, LTER sites within these systems could serve as early warning sites for the rest of the LTER network. Halfpenny also stressed the importance of modeling, and agreed with Garcia-Novo that historical and paleoecological data are useful in predicting responses to changing climate. Scientists watching the broadcast at the Bonanza Creek LTER also stressed modeling, and recommended the use of transects along climatic gradients and across ecotones.

Mark MacKensie (Temperate Lakes LTER, University of Wisconsin), Bob Gardner (Oak Ridge National Laboratory), and Judy Meyer (Coweeta LTER, University of Georgia) noted the total absence of aquatic systems from the three working group reports, and pointed out that aquatic systems are important parts of the global carbon budget because they store or release large amounts of carbon, especially CH₄, which is a potent greenhouse gas.

David Greenland (Niwot Ridge LTER, University of Colorado) suggested that more careful thinking needs to be done concerning the way in which climate and ecosystem change is studied. He was in favor of letting the ecosystem define the time scale of change rather than imposing conventional statistics of climate change on the ecosystem. Greenland called for new indices of climatic variability, and gave as an example changes in dominance of various air masses moving across continents.

More information on data management and on-line access to data was requested by a number of respondents. Kari Laine described the Inter-Nordic efforts in long-term monitoring, in which each country has a program director and a small group of scientists who act as the steering body for that country's monitoring. All sites send results to a data handling center in each country. On-line access between sites and from sites to the data handling centers is under development. Scientists at the Konza Prairie LTER (Kansas State University) were impressed that the reports all suggested a balance between modeling, experimentation, and monitoring, but believed that a concurrent need to pull these components together through innovative data management was underemphasized. Don Strebel (Versar. Inc., NASA) agreed that this was a good comment. He then described four different types of data management systems: (1) that which makes raw instrument data usable. (2) a data management system that supports an active field experiment. (3) operational support of a data base or archive beyond the field stage, and (4) deep archives or self-contained data bases. During the active experimental stage it is important to use on-line data bases which are connected through a network. Scientists can then proof data and work collaboratively with them. At this stage, the data system becomes part of the scientific research, as well as determining quality assurance. Such a system involves many non-standard approaches, and requires interactive development technology which will include scientists, be iterative, and will allow the system to evolve as the science emphasis changes.

John Vande Castle (LTER Network Office. University of Washington) wanted discussion of the information content of remote sensing data. how it relates to the kinds of long-term ecological research proposed, and what kind of archiving and distribution methods are necessary. Strebel responded that information varies with the scale of the remote sensing data used. Advanced very high resolution radiometer (AVHRR) data has a resolution of 1 km and shows the broad character of a region at a scale larger than any single LTER site can show. Higher-resolution data can have a resolution of 10-20 meters, which allows examination of a given stand of trees, for example. This scale of information is very important in dealing with levels of change over a long period. Through the capability of remote sensing, a long-term historical archive is now being accumulated that will eventually allow scientists to know what change occurred in a given stand of trees over a period of 10 to 20 years. It is also important to build an integrated data base at each LTER site based on extant data bases of all kinds, including remotely sensed images. Data screening is necessary to produce a distributable and usable amount of data. Strebel gave as an example a five-fold reduction in data that occurred at the FIFE experiment at the Konza Prairie LTER; 100 gigabites of remotely sensed data were reduced to 20 gigabites of usable data.

Partially in response to the comments of Caroline Bledsoe of the National Science Foundation (NSF) that common experiments which could be performed across global LTER sites seemed to be missing, the focus shifted back to the three working groups. During the break between the first and second programs, each group had discussed just such cross-LTER experiments, which they decided to call flagship experiments. The temperate forest group wanted to develop a Forest Ecosystem Manipulation Experiment (FEMEX) or Whole Ecosystem Manipulation Experiment (WEMEX). This experiment would entail placing a controlled environmental chamber over a forest to examine whole-system responses to enhanced temperature, moisture, and CO₂ levels. The group acknowledged that while this type of experiment is expensive, it is needed to define the changes in response to global warming. The second type of experiment is to measure the fluxes of the radiatively active gases at a given site, not just instantaneously but consistently over a period of time at a number of sites. Heal called this sophisticated measuring apparatus a RAGRIG (rig for measuring the radiatively active gases). The RAG measurements should not be done in isolation, however; a suite of experiments examining the biological processes producing the gases should first be performed in order to predict what is to be measured by the RAGRIG. The combined data should then be integrated with remotely sensed data. This integration of different scales of collected data will provide the opportunity for validation of predictions and for expansion and extrapolation to other sites. The third experiment the temperate forest group recommended was a Biology of Forest Growth (BFG) experiment to produce maximum growth at a given site to understand the maximum carbon uptake of a given system under conditions of enhanced CO₂ levels world wide.

Small-scale experiments are important to enhance the information gained in the big flagship experiments. These small experiments may not be very exciting in isolation, but when performed in concert with the big experiments across sites, they become powerful means of comparing different sites. Suggested experiments included forest floor warming and wetting, soil organic matter-removal or addition of litter, and transplantation of organisms over a range of sites to allow better prediction of species' responses to new conditions they may encounter as global warming proceeds.

The arid/semi-arid zone group determined some rules of water manipulation, including real-time addition or removal of precipitation, i.e. to lengthen a natural event by continuing to water after the storm or to shorten the period of precipitation by introducing the rain-out shelter before the natural storm ended; water manipulation based on rainfall intensity, time interval, length, or time between intervals and seasonality of precipitation; and water manipulation based on simulating historical climate by augmentation or reduction of water. These rules could be based on calculated means and variances of historical precipitation records from such sources as tree ring analysis. An important constraint on the experimental design is the choice of a precipitation regime which will not generate runoff if the topic of interest is soil water storage and water infiltration. The quality of water used to augment a precipitation regime is also an important consideration. Two kinds of sites could be used for these experiments: sites with different known histories of human impact, or those where different human impacts can be simulated at a given site. The arid zone working group considered fertilization experiments, but decided that manipulation of water should be the first close focus of experiments in the world's arid lands.

The boreal/tundra zone group wanted to stress the research opportunities in each ecosystem type, believing that one of the prominent characteristics of this system was that of gradients from tundra underlain by permafrost across the tree line into boreal forest, and on into the temperate forest. The group suggested several sites or nodes where intensive experimental manipulation had been done, such as fertilization of plants and the effects on plants of air and soil temperature, moisture, herbivory, fire, and CO₂ addition. Gus Shaver thought that the controlled environment experiments might work well in tundra because of the low stature of the plants growing in the tundra system. Measurements here could include changes in such state variables as biomass, controls on mineralization, or changes in process after 10 years of fertilization. These changes could be followed by remote sensing. Between the experimental nodes lined up along a gradient, climate and microclimate variables should be recorded, and nutrient levels in soils sampled with resin bags or buried bags. Decomposition and mineralization measurements would use litter bag experiments. There should also be some land/water manipulation.

The group responded to comments from watching LTER scientists who wanted to know why the group believed a georeferenced data base was important in the boreal/tundra system in particular. Shaver answered that it was important to know where these systems are, and how much of the earth is in this condition, so that the contribution of the boreal/tundra system to global elemental cycling and the carbon budget can be understood. The group was also asked if there was a need for them to work with non-biologists-hydrologists, geologists, etc. There was agreement that soil chemists, paraglacial geologists, and hydrologists were vital collaborators in understanding this system. The phenomena of glaciation, permafrost, and the permafrost/soil moisture relationship are poorly understood, and exert strong controls over biological processes, particularly those of the tundra.

Steve Storch discussed the technology available for communication within the proposed international research network. Storch had previously given a presentation to the members of the Albuquerque workshop. He pointed out that telecommunications are essential for such distributed research efforts. Although this is a new way of working for ecologists, interactive communication of the kind which will be required to build an international research network has been used well in other scientific disciplines. He mentioned two kinds of tools: the interpersonal or group interactions which are one-on-one in real time, and the international computer-to-computer network infrastructure. These tools must be approachable, and must be documented and supported by a technical staff experienced in communications technology and in helping the user to conceive of creative uses of the technology.

Paul Risser concluded the second hour's presentation by reminding all participants that the time is unique in the history of the earth and offers unique opportunities for our understanding of the functioning of the globe. In the face of global warming and the effects of human impact, not only are we offered a unique responsibility for managing an individual forest, portion of tundra, or expanse of arid grassland, but we are confronted with the opportunity to manage the sustainability of the entire globe. In the second workshop, Paul Risser continued, important and specific research topics have been identified, and projects, small to large, have been suggested for research. A single experiment will not work to answer the questions posed at this Albuquerque workshop; these questions demand a plurality of approaches, for which three steps must be followed:

1. As a minimum, the workshop should stimulate ideas for further proposals in the larger circle of scientists doing long-term research;
2. Proposals from the workshop should define programs and convince funding agencies and governments that the research needs to be done; and
3. Entire programs, identified in the workshop, must be funded so that the research can go forward.

The second television broadcast ended with an open discussion by the workshop participants about the needed research. Bill Heal urged everyone to think of how groups should be put together. In Europe the land area is small, but the differences between people are large. There is funding right now, however, for integrative projects in Europe. Peter Beets also thought the time is right to press for long-term research in New Zealand; the government is presently placing a strong emphasis on multidiciplinary research in highly relevant programs, especially those involving global change. Mary Seely noted that there is growing communication across political entities in southern Africa, and that biome projects are reorganized. Here too, the time is propitious for collaborative research. Uriel Safriel urged everyone to remember that sound research proposals are important if funding is to be forthcoming, especially in smaller countries that would be flattered to be included in collaborative projects which the global community believes essential. Francisco Garcia-Novo ended the discussion by suggesting that this effort will require large amounts of money and a lot more co-ordination than it has at present. He thought that great attention should be placed on land use and suitable management for conservation. Global co-operation in research also requires global co-operation to provide funding.

When the televised portion of the workshop ended, the formation of an international community of collaborators and research projects seemed to have been established; a valuable achievement, provided that international communication, momentum, and funding continue.

14.5 RETROSPECTIVE ON THE VALUE OF THE ALBUQUERQUE WORKSHOP

 Three months after the Albuquerque Workshop, a questionnaire was sent to all workshop participants and to scientists at each LTER site receiving the televised portion of the workshop. The purpose of the questionnaire was two-fold. First, it was important to find out what was happening in the development of international, collaborative research on global-scale problems as a result of the workshop; and second, it was important to give both those in the audience and those who participated in the workshop a chance to reflect on what had or had not happened since the workshop. The three questions on the questionnaire were:

1. Did the telecommunications workshop influence your thinking about global-scale problems and international co-ordination of long-term ecological research?
2. Did the workshop influence the research you are doing or will do? Did it suggest research questions you are exploring, techniques you are using, or contacts in other parts of the world?
3. Did the workshop and/or television broadcast stimulate your collaboration with other researchers to study phenomena of global warming which may be expressed differently in various habitat types? Are you involved in any international collaborations that grew out of this workshop?

Responses ranged from descriptions of collaborations among a few individuals to international efforts to approach and study global-scale ecological problems. The workshop inspired some researchers to develop projects that would give students experience in the co-operation among different disciplines required to tackle global-scale ecological problems. The workshop also generated a thesis which used historical records to examine the record of climate anomalies in Namibia.

Within countries, collaborative research is being nurtured. For example, the issue of global climate change has been introduced into all regional and local planning in which the Nature Reserves Authority in Israel is involved. Areas where there is a steep climatic gradient seem ideal for exploring the effects of warming, and also the independent effect on productivity of increased carbon dioxide concentrations. Desert meteorologists, plant environmental physiologists, and desert ecologists are reported to be collaborating along such a gradient in Israel.

For researchers in Third-World countries, the workshop reinforced the perception that much of the large-scale research on global problems is financially out of reach for these countries. An important addition to a few large flagship experiments are small-scale, inexpensive projects to be conducted at many places on the globe. Smaller efforts stimulate thinking and training, and have a potential for local funding. Examples of some of these efforts are a collaborative project to study surface-active faunas of African, North American, and Central American arid areas using similar techniques of data collection and analysis; and testing ideas concerning multiple origins of weather affecting interannual rainfall variations seen in the arid South-western United States and in Africa. Collaborative work at the Sevilleta and Jornada LTER sites in New Mexico and at Mapimi, Mexico, were also reported. Research themes there are (1) variability of the influence of the ENSO (El Niño Southern Oscillation) phenomenon along the Chihuahuan Desert latitudinal gradient, (2) importance of abiotic decomposition of litterfall along the same gradient, and (3) experimental studies of the variability of plant genotypes from different geographic origin.

It is paradoxical that more collaborations were reported for the arid/semi-arid biome than for the boreal forest/tundra biome which was singled out by one respondent as more amenable to global-scale research than either the arid/semi-arid or temperate forest biomes. The physical continuity of the boreal forest/tundra biome should produce more uniform responses to global changes across the biome,  whereas the varying surrounding environments of the arid/semi-arid and temperate forest biomes should make them sufficiently dissimilar that they would not show uniform responses to global changes.

Global-scale problems require global networking as an essential methodology. Several respondents believed that the creation of financial and other resources had to precede the formation of international links. Just as within each LTER site there is 'networking' between sampling, monitoring, and experiments, and between the specialists that operate them, so must there be similar networking between various sites on the globe.

Co-ordination of research and effective international communication are necessary in the areas of (1) site selection, output objectives of LTER, and parameters to be measured to ensure global representation of key processes; (2) co-ordination of research program development to ensure that core areas are covered to the extent and precision agreed upon; and (3) analysis and synthesis, so that the research and monitoring programs at LTER sites are based on up-to-date information. Innovative communication is essential to maintain the network of international research, and international co-operation will be especially valuable in the exchange of techniques and models. One participant said the workshop convinced him that solving ecological international problems requires international teams of scientists and others in addition to new ways of co-ordinating such an effort. Sophisticated telecommunications, such as that demonstrated by the workshop, are required, and the respondent is designing computer modeling programs to handle global-scale problems and to foster international co-ordination. It was suggested that long-term ecological research should be a partnership between researchers, industry, government policy developers, and the public. The involvement of all these groups will encourage co-operation arising from the improved understanding of the global nature of environmental issues. Information exchange should be done so that the real costs of resource management are clearly evident. The implications of global policy at local and regional scales are far reaching, and research results need to be convincing before governments are likely to agree to act. Presently, the governments of New Zealand, the United Kingdom, and Australia have agreed to joint climate change research.

The television broadcast and the responses from the audience indicated a willingness by the wider scientific community to get on with the job. New Zealand in particular shares this desire to do something. An international LTER effort could be splintered, however, by the individual initiatives emerging in different countries unless a truly co-ordinated effort is made to integrate them. An overall co-ordinating group must be formed now to manage the several research proposals under consideration. Other international initiatives, such as IGBP's GBO network headed by Rafael Herrera, are underway. It was suggested that the international network of long-term research groups should be represented on the IGBP international committee.

In summary, workshop participants reported the beginning of research collaborations between individuals and between sites in a given country , and plans for co-ordinated research within countries. It seems clear that the components for internationally co-ordinated long-term research at a global scale are present. The next task is the assembly of these components and the use of sophisticated remote sensing, geographic information systems, and both person-to-person and computer-to-computer communication, if there is to be true collaborative, co-ordinated, and international ecological research.