1. Introduction
     To explore the opportunities for research on climate variability in the Americas from high elevation ice cores, a group of 20 scientists representing 11 countries met in San Carlos de Bariloche, Argentina from 11-13 December, 1996. (participant list with group photo) Discussions focused on regions of the Americas from which additional ice core records could be recovered, and the value of each location in terms of understanding climatic variability. The geographical scope of the workshop discussion extended from the Antarctic Peninsula (~ 74°S) to northern Ellesmere Island (83°N). The meeting was financially supported by the Inter-American Institute for Global Change Research (IAI). The primary goals of the workshop were to identify both the principal scientific questions which can be answered, and societal issues which can be addressed, by ice core studies along a North-South transect through the Americas, to determine which ice caps have the potential of yielding high resolution paleoenvironmental records, and to initiate discussions on programs of collaborative research.
    The first session of the workshop was devoted to presentations reviewing the current state of knowledge in three areas:
1. Circulation regimes of South America and meteorological observations at high elevation;
2. Methodological techniques and issues of ice core analysis; and
3. Regionally-specific research accomplishments related to glacier-climate interactions and ice core drilling in the Americas.
Discussions were then conducted as a group on individual regions of the Americas where there are excellent prospects for recovering paleoenvironmental records from ice cores. On the third day, we split up into small groups, each with a regional focus, to discuss specific needs and ideas for research projects in each region. The group then came together once again to hear summary presentations from the small groups and discuss integration of the projects into a coordinated proposal.

2. Rationale
    Whatever anthropogenic effects on climate occur in the future, they will be superimposed on a background of natural climatic processes which may mask, or magnify, such impacts. To understand and anticipate what lies ahead, or even to understand our contemporary environment, requires an assessment of conditions which preceded the present, a perspective which can only be gained by studies of the past. Such research focuses on both the time-evolution of relevant processes (which may operate on time-scales from many millennia to a few decades) and on their interactions through time. This may shed light on the linkages between biogeochemical processes and the physical aspects of climate, enabling causes and effects to be isolated and assessed. Studies of the past also reveal how quickly earth systems have responded to particular forcing factors, which is of critical importance in anticipating and planning for future environmental changes. Finally, a better understanding of environmental conditions in the past is invaluable for testing numerical models of atmospheric and oceanic circulation, biospheric dynamics and environmental processes, by providing a database of environmental conditions in the past which were quite different from those of today (Bradley and Eddy, 1991).
    Climatic variations of the last 2000 years are of particular relevance to contemporary human endeavors. Although climatic fluctuations over this time period were small in comparison with those expected from a doubling of greenhouse gases (e.g. generally <2°C in decadal mean, seasonal anomalies for any one region, versus 4-8°C expected in the 21st century) the last 2000 years have witnessed the most extreme conditions of the Holocene period. Furthermore, there is abundant evidence that even such relatively small climatic anomalies have had significant impacts on human societies, for example, as a result of prolonged droughts, and other climate-related disruptions (Thompson et al., 1988). We must understand these fluctuations and their causes (forcing mechanisms) and the interactions (teleconnections) between one region and another because it is on this background of natural variability that the anthropogenic greenhouse effect is being imprinted. Instrumental records are too short to provide the necessary perspective. Only climate-related "proxy" records can provide the longer time frame necessary to document the significance of underlying trends, to analyze the nature of any periodicities in climate, and to address the critical issue of "abrupt" climate change (Thompson and Mosley-Thompson, 1987). Temperatures today are higher than they were in the late 19th century, but we can not yet say definitively how much of the observed warming is due to natural oscillations of climate, and how much is related to human activities (Bradley and Jones, 1995). Long-term, high resolution proxy records from poorly-studied regions of the world (e.g. Latin America) are an essential pre-requisite to placing contemporary climatic conditions in perspective, and constructing a globally comprehensive view of climatic variations in the past.
    Ice cores provide excellent, high-resolution records of past environmental conditions and climatic variability. Along the western margin of North and South America are a nearly continuous series of ice-capped or glacierized mountains, containing a unique archive of information. Annual layers of precipitation record the atmospheric processes involved in moisture transport to the ice caps (e.g. Grootes et al., 1989); particulates and dissolved material deposited with the snow document atmospheric constituents; gases trapped in the ice record past atmospheric composition. Geochemical studies of ice cores provide a comprehensive view of prevailing air masses and of atmospheric circulation in the past (e.g. Mayewski et al., 1993a). Some of the most important discoveries about earth system processes have been obtained from ice cores from high latitude ice sheets (e.g. Jouzel et al., 1989; Dansgaard et al., 1993; Taylor et al., 1993). However, these studies can not address fundamental questions of significance to lower latitudes; in particular, climate variations related to ENSO are of great significance in the Tropics, accounting for much of the variance in temperature and precipitation in this region (e.g. Bradley et al., 1987). Ice cores from low latitudes generally have high accumulation rates, providing high resolution, direct records of precipitation and temperature changes due to ENSO (Thompson et al., 1984a, 1985). Together with other proxy records, long-term changes in ENSO variability can be assessed (Baumgartner et al., 1989). The importance of long-term data on ENSO from paleoclimatic time series was explicitly recognized by participants in the IAI Workshop on ENSO and Inter-annual Climate Variability (IAI, 1995). Additional ice cores from low latitudes of the Andes should help in further documenting past ENSO activity, by constraining uncertainties in the few existing records. No other high resolution proxy records of past climate are yet available for the inter-tropical Americas.
    ENSO variability is not the only important quasi-periodic variation observed in long-term climate records. Recent studies have demonstrated that lower frequency variations (10-20 years), possibly associated with other large-scale climate systems (such as the "Pacific-North America" or PNA circulation mode) play a role in decadal-scale climate variability in the extra-tropical regions of the Americas (Mann et al., 1995). By obtaining more high resolution ice core records along the western axis of the Americas, from low to high latitudes, it should be possible to shed further light on this aspect of climate variability, and to what extent these are linked to lower frequency aspects of the ENSO system.
    Recent observations have clearly documented the urgency of recovering ice cores from high altitude sites in the tropics. Ice caps and glaciers in the region are rapidly melting and retreating, at a rate that threatens to destroy unique paleoenvironmental records (Hastenrath and Kruss, 1992; Schubert, 1992; Thompson et al., 1993). Freezing-level heights have been rising at low latitudes, apparently in response to decadal-scale changes in tropical sea-surface temperatures (Diaz and Graham, 1996). There is also the suggestion, based on general circulation model experiments, that high elevations in the Tropics may be particularly vulnerable to greenhouse-gas induced warming (Mitchell et al., 1990), providing additional pressure to develop a coordinated, international high altitude ice core research program. At the same time, models project that high latitudes will ultimately witness the largest changes in temperature. A transect of ice cores from high southern latitudes (Chile/Argentina) to high latitudes of the northern hemisphere (Alaska/Yukon Territory) would provide a valuable perspective on the underlying climate variability throughout the Americas, and on inter-hemispheric linkages in the timing and mechanisms of climate change, at a time when greenhouse gases are rising exponentially.

3. Region-specific Workshop discussions:  promises & prospects
    At the workshop, five regions of the Americas where new ice core records might be obtained were delineated, as a framework for discussion:
· Antarctic Peninsula
· Patagonian Icefields
· Andean ice caps and glaciers between the Patagonian Icefields and the Dry Axis
 (~ 46° to ~28°S)
· Inter-Tropical Zone
· High latitudes of North America.
Group discussions concentrated on the scientific questions that could be addressed by ice cores from each region. Previous ice core, glacier, and climate research within each region was also discussed, including geophysical surveys, short ice cores, mass balance programs, and meteorological data. In addition, the group attempted to address the uncertainties, special considerations and/or issues unique to each region. These discussions were designed to provide the background for development of strategic recommendations.

3.1 Antarctic Peninsula
    The Antarctic Peninsula begins less than 1,000 km from southernmost South America, and extends south from ~ 62°S, to where the peninsula merges with the continent at ~ 74°S. There are major ice shelves adjacent to the Peninsula, the break-up of which would have global implications. Indeed, the demise of several minor ice shelves (Doake and Vaughan, 1991; Rott et al., 1996) supports predictions about the climatic sensitivity of ice shelves on the Peninsula (Mercer, 1978), and has created a sense of urgency to understand past climate variability on the peninsula (cf. Fahnestock, 1996). Several existing ice core records (see for example Aristarain et al., 1990; Peel, 1992) have already been helpful to the understanding of climatic variability in the region. These records may also contain evidence for ENSO variability in the past. In addition, due to the oceanic proximity, records from the peninsula are likely to contain an important record of marine biological activity, as existing ice core records and the University of São Paulo aerosol sampling program has shown. Ice cores have also produced information about local volcanism (Aristarain et al., 1982).
    Considerable data collection is currently being conducted on the peninsula, relative to the other potential regions the group considered. For example, in the Antarctic Peninsula region there exist the oldest and largest number of meteorological stations. Also, the British operate at least five automated weather stations (AWS) and seven are operated by the North Americans (University of Wisconsin; only two of these are at elevations above 100 m: ‘Uranus Glacier’ at 780 m and ‘Ski-Hi’ at 1395 m). From these data and the manned stations, it is known that there are large east-west gradients of temperature and precipitation, which will complicate the spatial extrapolation of results from single ice-core sites. In 1997, an Argentine-French project on the Peninsula will commence drilling to a planned depth of 300 m, which will hopefully cover 2,000 years.
    Despite the prior and on-going work on the Peninsula, discussion emphasized that additional efforts will be required to understand the isotopic signal in the cores. This would involve increased sampling of precipitation, and experiments tracing the isotopic concentration of the snowpack through time. The AWS data will require careful analysis to explore variability in precipitation source area, which seems likely given the oceanic proximity and interannual variability in sea-ice cover. Meteorological modeling support would be essential to working with these data, and the Brazilian group has begun such efforts.

3.2 Patagonian Icefields
    The North and South Patagonian Icefields are two relatively narrow ice masses stretching 100 and 360 km, respectively. The North Patagonian Icefield (NPI) is approximately one-third the total area of the South Patagonian Icefield (SPI), and the mean surface elevation is approximately 500 m lower (Warren and Sugden, 1993).
    Ice core records from Patagonia would provide basic information on precipitation at high elevations in the southern Andes, about which very little is known. The combined latitudinal extent of the two icefields provides an opportunity to investigate changes in the position and strength of westerly airflow through time. As on the Antarctic Peninsula, there is a strong east-west climatic gradient across the icefields, the strength of which has also probably varied through time. Ice cores from a network of carefully selected sites could help resolve both of these questions.
    A limited number of investigations have been carried out on the accumulation areas of the icefields (e.g. Yamada, 1987; Aristarain and Delmas, 1993), despite considerable recent work by Naruse, Skvarca, Aniya, Casassa and others on the marginal position (extent) and thickness changes of Patagonian glaciers. This work in the ablation areas indicates that most NPI and SPI outlet glaciers have been retreating (Aniya, 1988; Aniya et al., 1992; Naruse et al., 1995), although others such as the Moreno Glacier seem stable (Skvarca and Naruse, 1997). Measured surface lowering rates in the region, such as 10-15 m/yr for the Upsala Glacier in the early 1990s, are among the highest in the world (Naruse et al., 1997). In the accumulation areas, Casassa has recently had a student working on the Chico Glacier (SPI, north of Fitzroy), and Naruse has been overseeing work on the NPI accumulation area which has included a 14.5 m deep firn drilling at 1500 m a.s.l. in December 1996. They, a team of Japanese graduate students, failed in radio echo sounding, because an oscilloscope placed on a sledge was blown away by winds!
    Patagonia presents at least three special conditions, relative to ice core drilling. First, the region is relatively warm. Using measured temperatures at low elevation and assumed lapse rates Naruse estimates that the 0°C maximum temperature level is ~ 2500 m, which is higher than most of the accumulation areas. Melting, and ice layers, are therefore to be expected in any core from the region. Second, the high precipitation on the icefield suggests that very deep drilling will be required to extend the ice core record back a reasonable time, however, net accumulation rates are also unknown. In one location — at 2680 m in the accumulation area of several large glaciers in the SPI — a 13 m core indicated an accumulation rate of only 1.2 m/yr water eq. (Aristarain and Delmas, 1993), although Naruse et al. (1995) estimate that 6-8 m of accumulation in water should be required to support the Moreno Glacier. Lastly, field work in the region is extremely difficult, due to wind and precipitation. Aniya et al. (1996) note that nearly cloud-free images from three adjacent Landsat scenes are available for only one day in nearly 25 years of data!
    The dearth of field work in the accumulation area suggests a need for reconnaissance geophysical surveys, short cores recovered from the highest elevations, and installation of AWSs. Geophysical work could provide a picture of basic bedrock topography as well as the glaciological information currently lacking, which will be necessary to select the best drilling sites. However, any geophysical techniques employed must be capable of dealing with liquid water within the glaciers, possibly even at the highest elevations. One drill hole made at 1300 m on the San Rafael Glacier to 38 m contained a two-meter thick aquifer on the firn-ice interface at about 25 m in depth (Yamada, 1987). Due to the high accumulation rate on the icefields, there is a need for a network of medium-length cores (~ 50 m), which must be drilled with lightweight, portable equipment suitable for use in the remote location. This effort would provide a useful indication of spatial variability in isotopic and geochemical concentrations, and allow measurement of temperature profiles. Detailed east-west and north-south transects of short, hand-drilled cores should also be considered prior to efforts to select new deep drilling sites. Although there are some meteorological station data from locations such as Lago Argentino (1940 to present) and Punta Arenas (1870 to present), the sharp climatic gradients in the region render these data difficult to extrapolate spatially; Naruse has measured such a sharp gradient over a distance of only 60 km (Takeuchi et al., 1996; Naruse et al., 1997).

3.3 Patagonia to the Dry Axis
    Important glacierized zones extend along the Andes from north of the NPI to Cerro Potro (~28°S, 5864 m), the most northerly massif that is out of the Dry Axis. Included among these is the 300 km2 Aconcagua massif (~ 33°S; 6960 m). This region also contains numerous active volcanoes which are concentrated between 32 and 42°S. Glaciers and rock glaciers are very important in the region as water resources, particularly on the drier eastern side (Argentina). These mountains come under strong influence of westerly circulation during the winter, resulting in a sharp west to east precipitation gradient (as in Patagonia). In contrast to mountains north of the Dry Axis, the principal moisture source for glaciers is from the Pacific Ocean to the southwest. Ice core records from suitable sites are likely to contain a strong ENSO signal.
    Previous meteorlogical observations have occurred on Maipo (34°S and 69°50’W; 5200 m), where a weather station existed at 3700 m for a few years, allowing an estimated maximum temperature of -9°C to be calculated for a potential drill site at 5200 m. Near Tupungato volcano (32°20’S and 69°45’W; summit at 6550 m), where a mass balance study has been carried out by Escobar since 1975, a few potential drill sites exist at ~ 5,000 m, where the estimated mean annual temperature is about -11°C. One of them, a glacier on the "Mesón San Juan" mountain (~33°33’S and 69°46’W, 6035 m), was approached by Aristarain in March, 1997 with drilling equipment. Unfortunately, it was not possible to accomplish the mission due to large penitentes and crevasses not found in previous years. Further north, there is another potential drilling site on Co. Tapado (5536 m, 30°08'S, 69°56'W) in the Elqui Valley. Compared to a weather station at 3100 m, the estimated mean annual temperature is between -7 and -8°C, while the 0°C isoline is situated at 4300 m. The Co. Tapado and the Co. Potro site will probably be tested with short ice cores during summer 1997/98 (University of Bern, SNF project).
    There are very local climatic effects around the existing meteorological sites in the region, Aristarain reported. As a result, temperature estimates may be in error, and some sites may not be high enough in elevation to avoid significant meltwater percolation. (Indeed, a d18O elevational profile made by Holdsworth, from snow samples he collected en route to the summit of Aconcagua, suggests that coring there would have to be done above 6200 m, if an ice cap existed.) At other sites, particularly those that are not ice caps, there may be flow disturbance to the glacier stratigraphy induced by underlying topography. This would limit the useful length of any records recovered.

3.4 Tropics
    Potential ice core drilling sites exist in the inter-tropical region between Nevado Sajama (18°S) just north of the Dry Axis, and the equatorial volcanoes of Equador. There is a great need to explore these sites for new records, as throughout the region glaciers are retreating (e.g. 160 m at Antisana this century) and evidence from multiple sources indicates that freezing-levels are rising in the tropics (Diaz and Graham, 1996). Water resources will become increasingly important in the region.
    The scientific questions that could be addressed by additional ice cores from the tropics are numerous, as demonstrated in previous coring efforts by Thompson. Potential sites for long records are relatively abundant. Questions that could best be addressed by a network of records include:

• How has ENSO varied in the past? This is an enormously important societal issue; in Ecuador, for example, ENSO events induce much greater than normal annual precipitation, and the resultant flooding has a great impact on human activities (600 deaths in 1983). Contemporary mass balance investigations in Bolivia and Ecuador indicate that tropical glaciers of the Americas respond strongly to ENSO variability (Pouyaud). Ice-core records from the Tropics contain a strong ENSO signal, due to its influence on precipitation.
• What is the relationship between deforestation in Amazonia and precipitation in the Andes? What is the past record of biomass burning (the trajectory of air influenced by biomass burning crosses many potential sites), and are changes in vegetation reflected by pollen deposited on the ice, as the Huascarán core demonstrated (Thompson et al., 1995)?
• Why are tropical ice caps and glaciers disappearing so quickly, and how is climate changing as a function of elevation?
• Can a double annual signal be resolved in the equatorial records, from Amazonian and Pacific air masses? Sites in Ecuador contain a double peak in precipitation each year, with the maximum precipitation derived from the west side in March and April, from the east side in June. If the two periods can be resolved in the ice cores, a double signal should be available.
• Lastly, does analysis of 10Be in tropical ice cores provide a useful proxy for earthquake events of the past? This is a seismically active region, and dust stirred up by mass movements is incorporated into the lower troposphere and redeposited on glaciers in the region.

3.5  North America (Canada and Alaska)
    Discussion of potential sites within this vast region concentrated on the St. Elias Mountain Range on the border between Alaska and the Yukon Territory, the Alaska Range of interior Alaska, and the ice caps of the Canadian Arctic Archipelago, within the Nunavut region. The potential and existing Alaska-Yukon ice core drilling sites are relatively high elevation (e.g. Mt. Logan at 5330-5340 m), and most are located just south of the Arctic Circle (e.g. Mt. Logan at 60.5° N). The climate at these sites is severe, as evidenced by a 10 m temperature of -29.5°C recorded in the PR col between sub-peaks of the Mt. Logan massif (PR col refers to that between Prospectors Peak and Russell Peak). The northern Canada (Nunavut) sites are at considerably higher latitudes (75° to 83°N) and all are less than ~ 1700 m elevation. As a result of these differences in geographic location and elevation, the scientific questions that the two regions can address differ somewhat.
    Although the airflow is predominantly westerly in the Alaska-Yukon region, and air masses of maritime Pacific origin are dominant at low elevations to the south and west of the mountain ranges, the potential ice core drilling sites are above the level of both marine and continental aerosol influence (Holdsworth and Peake, 1985). As a result, the geochemical signals in the ice core records should be reflective of mid-tropospheric atmospheric composition. These sites thus record background atmospheric dust levels, due to their isolation from local sources (Mayewski et al., 1993b). Indeed, there was no anthropogenic SO4- 2 signal in the Mt. Logan core (Monaghan and Holdsworth, 1990). Additional aspects of climate variability that could be addressed at Alaska-Yukon sites include the Pacific-North America circulation (PNA) pattern and the influence of ENSO on precipitation. Due to the prevalence of active volcanoes, particularly in the Aleutian Islands and north Pacific rim, these ice cores have the potential for excellent chronological control (cf. Holdsworth and Peake, 1985).
    There have been two deep (> 100 m) ice cores recovered in the Yukon: at 5,300 m on Mt. Logan in 1980 (G. Holdsworth, 103 m; adjacent cores of 46 and 51 m), and at ~3000 m on Eclipse Dome in the St. Elias Range in 1996 (E. Blake, 160 m of possible 500 m depth). The Mt. Logan record covers ~ 300 years, at a site where net annual accumulation (snow) is ~ 1.1 m. The record from the Eclipse site contains a higher proportion of melt layers, due to the lower elevation. This site has a 2-year meteorological record. There have been other short to medium-depth drilling efforts in the region, including a 40 m core from Mt. Wrangell in Alaska taken in 1984 (although this was strongly contaminated by H₂SO₄).
    At potential ice core sites in the Alaska-Yukon region, geophysical surveys will be especially important prior to selection, because of glacial flow considerations. Topographic relief is high, and these glaciers have a high mass turnover, requiring a large flow divide area to obtain ice which is not sheared at depth. Topography greatly influences the distribution of precipitation as well, because high velocity winds create areas of scour and deposition. Snow accumulation at potential sites must be representative as possible of regional precipitation, despite variability of wind direction.
    The northern Canada ice cores have been especially useful in the reconstruction of summer temperatures, through the analysis of melt proportions (Koerner, 1977). From the isotopic composition, mean annual temperature has also been estimated. Microparticle and the concentration of different chemical species provides information about high latitude atmospheric composition. Ice cores from the most northerly of the potential sites could also provide a valuable record of sea-ice cover on the Arctic Ocean. Although most sites suitable for ice core recovery experience melting during the summer, there may be localized cold sites where the records will be less influenced by meltwater percolation.
    Ice core drilling in the Canadian High Arctic began with the Meighan Ice Cap in 1965, the record from which consisted mainly of superimposed ice due to excessive melt, and which primarily reflected local conditions. The Devon Island Ice Cap was drilled in 1971 (as well as 1972 and 1973), and provided a record extending back > 100 kyrs that is well correlated with the records from Greenland. Another effort to recover a pre-industrial revolution record from Devon Island is planned for the spring of 1997. On the Agassiz Ice Cap of Ellesmere Island, six cores to bedrock have been drilled since 1977 (all between 1670 m (A77) and 1730 m (A93)). While the record from the summit was excessively affected by wind scour, another site 1 km lower has provided an excellent record. Two different sites have recently been drilled on the Penny Ice Cap of southern Baffin Island (333 and 180 m), and these cores are currently being analyzed.
    There is a need for additional spatial sampling surveys in both regions, to assess how variable the isotopic signal is. Such a program will be undertaken on the Devon Ice Cap this year in conjunction with the drilling. Meteorological monitoring is also needed at high elevation sites in these regions, as the only existing stations are on the Agassiz and Penny Ice Caps. The extreme winds and high snowfall will provide serious impediments to the operation of such stations in the Alaska-Yukon region. For example, at the PR col site on Mt. Logan an automated station produced nearly a year of temperature measurements in 1988, although the wind record is shorter because the anemometer was destroyed by high winds. The Agassiz weather stations demonstrate that riming and icing can render some sensors inoperable for extended period of time in the Canadian Arctic (B. Alt, pers. comm.).

4. Workshop Recommendations
    Workshop participants strongly recommended a program of ice core drilling at high elevations in the Americas. The potential regions for drilling projects differ widely in their geographic location, elevation, and climate. To assess the value of the different sites, and thereby evaluate their relative importance and feasibility, we identified several key questions to be addressed:

• What are the major environmental signals likely to be recorded in ice cores at each potential site and in each region?
• What is the mean annual temperature and maximum summer temperature at different potential sites in the region? The temperature regime will be an important indication of the extent of surface melting, which at some sites could be problematical. For example, a gas record (CH4, d18OATM) cannot be obtained if the mean annual temperature at the site is greater than -10°C, or if more than minor melting has occurred.
• What is the accumulation rate at potential sites in the region? This will be a determinant of the depth required for drilling, and the resolution available in resultant cores.
• If a specific site is indicated by the criteria above, will the record be regionally representative  (rather than solely reflective of local climate)?
• Is drilling at the site logistically feasible, and what are the costs of a drilling operation there?

    The alternative concept of drilling several deep ice cores would require a more focused investment of resources to a smaller number of sites. In some locations such a strategy may be crucial, as there are places (especially in the inter-tropical zone) where the rate of climatic change at high elevation is so great that if cores are not recovered in the near future valuable portions of the records will be lost. Such sites would be the first to be drilled. Details regarding specific sites suggested within each region are outlined below. Specific issues and considerations are also indicated.

4.1 Antarctic Peninsula
    There have been numerous cores recovered from the peninsula, and several additional studies are planned for 1998 by the British Antarctic Survey, and by the Instituto Antártico Argentino. A compilation and analysis of all of the data should be undertaken, especially for the recent past (e.g. since 1960), to aid in understanding the signals. There was some discussion that a couple of east-west traverses should be undertaken to acquire short cores, allowing a modern calibration. Despite numerous AWSs, calibration of the signal in ice cores from the peninsula remains a big issue. Strong support was voiced for the idea of compiling and analyzing all of the existing AWS and meteorological station data into a database. This work could be done as a joint project by students from South and North America. There is also a need to develop techniques for year-round aerosol sampling, to build upon the existing Brazilian program. The collection of precipitation samples also needs improvement, to better understand the climatic significance of isotopic variations.

4.2 Patagonian Icefields
    Prior to the drilling of a deep ice core on one of the icefields, additional investigations will be required. Information on bedrock topography, accumulation rates, thermal regime, and climatic gradients must be obtained to maximize the potential for obtaining a valuable ice core record. Due to the difficulties of fieldwork and logistics in the harsh climate of Patagonia (especially high wind and heavy precipitation, rendering aircraft support difficult), lightweight, portable equipment will be required for the geophysics and recovery of short cores. An additional recommendation was made to develop and build a lightweight drill capable of reaching depths of approximately 50 m, which will be necessary due to the expected high accumulation rate on the icefields. At present, little information exists on the firn/ice transition depth in the region. Also recommended were AWSs, perhaps installed on nunataks to minimize complications due to the high accumulation rate.

4.3 Patagonia to the Dry Axis
    Specific recommendations were not made, pending the reconnaissance of Tupungato in March. Nonetheless, some of the potential sites in this region are easily accessible, which would reduce the logistical expense of drilling. Ice masses in this section of the Andes are very important to the water resources of the region (especially for agriculture), and hence merit investigation in terms of their response to potential climate change.

4.4 Tropics
    Several possible sites in Ecuador emerged as high priorities for drilling in the future. The glaciers of Antisana (5758 m summit) are very important for water resources and hydroelectric power generation, and a recently obtained 10 m core will provide information about the suitability of the site. In addition, the Ecuadorian Atomic Energy Commission operates a station near the mountain at Papajata, which would be of great value in the calibration of the isotopic record. Cotopaxi would be another target, to obtain a record before the mountain erupts again. Prior to drilling the effect of heat flux from the mountain on the thermal regime of the glaciers should be investigated. Establishing additional AWSs at high elevations in Ecuador is also a high priority, to help assess the scientific issues that could be addressed by ice cores from the volcanoes, and permit calibration of any cores drilled; B. Francou of ORSTOM in collaboration with INAMHI have begun work on this.
    To the south in the Cordillera Real, Nevado Illimani (6439 m) would provide the best possibility of a good record, because reconnaisance short cores obtained from neighboring mountains by Francou (e.g. Mururata, 5868 m) indicate that signal quality improves with increasing elevation. A record from Illimani would also be useful in comparison with that anticipated from Nevado Sajama in 1997, as Illimani is directly upwind along the trajectory of moisture-laden air mass advection from the Amazon Basin, to the east.
    There is also a need to continue the existing monitoring programs and mass balance studies on glaciers in this region, as fiscal constraints seem to be reducing governmental interest and support (esp. in Peru). Lastly, it would be helpful to assemble the isotope precipitation data already collected, to further aid in the understanding isotopic signals at low-latitudes.

4.5 North America (Canada and Alaska)
    There were two sites identified as high priority locations for deep ice core drilling: on the col between Prospector and Russell Peaks (PR col) of the Mt. Logan massif, and on Mt. Oxford in northern Ellesmere Island. A reconnaissance of the PR col site was conducted in 1988-89 by G. Holdsworth who reported that the col is an area of roughly 400 by 400 m. He measured a 10 m temperature of -29.5°C, and installed an AWS which could not withstand the severe winds during the winter. Holdsworth estimates that the accumulation rate is currently about one meter per year (0.3 - 0.4 m w.e.). Radar data show that there is 200 m of firn and ice on the col, with a strong possibility that a record could represent the entire Holocene. The attraction to the Mt. Oxford site (2100 m) is that the record probably contains less melt than the Agassiz Ice Core, due to the slightly higher elevation and more northerly location. Additional sites within the Queen Elizabeth Islands could be evaluated based on extensive radar imagery flights over the area in 1995.

4.6 Other recommendations
    There was strong sentiment expressed by participants that collaboration resulting from the research should provide long-term benefit to the people and labs in Latin America. Direct participation in projects is desired, which could involve training in both field and analytical techniques, as well as basic technology, the goal being to "reduce the wide crevasse between north and south". Participants noted that while applied, practical applications of our science are needed to get governmental attention, there must also be support for basic research. This dilemma is illustrated by the recent privatization of ElectroPeru, which jeopardizes a long-running mass balance program in the country. Another specific idea was to have SENHAMI become more involved in operating and maintaining the AWS in the Cordillera Blanca.
    The need for training of Latin American scientists was a recurring theme, demonstrated by the fact that there are currently no professionally trained glaciologists in Bolivia, and Chile has only one, despite abundant glaciers and urgent water resource issues in both countries. It was suggested that IAI funds could act as a seed to establish or improve programs. This could be done through the direct involvement suggested above, and also through the concept of offering ‘short courses’ at various South American institutions. These courses would offer hands-on training by international teams of experts in different fields, who would reside at the institutions for short periods and be available for discussions.
    Realizing a program of research on climate variability in the Americas from high elevation ice cores would serve many goals of the IAI, because the program could be largely distributed in nature, and involve many different institutions in aspects that range from highly technical and specialized, to those which are more organizational in nature. Focused aspects of individual ice core projects could greatly benefit the countries involved by later transformation into studies with long-term monitoring emphases, such as the weather stations and glacier mass balance studies. All of this knowledge and information, along with these types of skills, will be vitally important to the countries involved as water resources become increasingly important. The knowledge gained about climate variability through a transect along the Americas will benefit a much larger region, as we attempt to better understand global climate change.

5. Acknowledgments
    This workshop was supported by a Start-Up Grant (Phase I) to the University of Massachusetts, from the Inter-American Institute for Global Change Research (IAI). We would also like to thank Ms. Soledad Luddeck and the staff of La Cascada for helping to make the meeting a success.

6. References

Aniya, M., 1988. Glacier inventory for the Northern Patagonia Icefield, Chile, and variations 1944/45 to 1985/86. Arctic and Alpine Research 20, 179-187.

Aniya, M., R. Naruse, M. Shizukuishi, P. Skvarca and G. Casassa. 1992. Monitoring recent glacier variations in the Southern Patagonia Icefield, utilizing remote sensing data. International Archives of Photogrammetry and Remote Sensing, XXIX, B7, 87-94.

Aniya, M., H. Sato, R. Naruse, P. Skvarca, and G. Casassa, 1996. The use of satellite and airborne imagery to inventory outlet glaciers of the southern Patagonia Icefield, South America. Photogrammetric Engineering & Remote Sensing 62, 1361-1369.

Aristarain, A.J., R.J. Delmas and M. Briat, 1982. Snow chemistry on James Ross Island (Antarctic Pen.). Journal of Geophysical Research 87 (C13), 11,004-11,012.

Aristarain, A.J., J. Jouzel and C. Lorius, 1990. A 400 year isotope record of the Antarctic Peninsula climate. Geophysical Research Letters 17, 2369-2372.

Aristarain, A.J. and R.J. Delmas, 1993. Firn core study from the Southern Patagonia ice cap, South America. Journal of Glaciology 39, 249-254.

Baumgartner, T.R., J. Michaelsen, L.G. Thompson, G.T. Shen, A. Soutar and R.E. Casey, 1989. The recording of inter-annual climatic change by high resolution natural systems: tree rings, coral bands, glacial ice layers and marine varves.  In: Climatic change in the eastern Pacific and western Americas, D. Peterson, (ed.), 1-14, Washington, D.C.: American Geophysical Union.

Bradley, R.S. and J.A. Eddy, 1991. Records of past global changes. In: Global Changes of the Past (ed. R.S. Bradley) University Corporation for Atmospheric Research, Boulder, p. 5-9.

Bradley, R.S., and P.D. Jones, 1995. Recent developments in studies of climate since A.D. 1500. In: Climate Since A.D. 1500 (Revised Edition) R.S. Bradley and P.D. Jones (eds.) Routledge, London, 666-679.

Bradley, R.S., H.F. Diaz, G.N. Kiladis, and J.K. Eischeid, 1987. ENSO signal in continental temperature and precipitation records. Nature 327, 497-501.

Dansgaard, W., S.J. Johnsen, H.B. Clausen, D. Dahl-Jensen, N.S. Gundestrup, C.U. Hammer, C.S. Hvidberg, J.P. Steffensen, A.E. Sveinbjornsdottir, J. Jouzel, and G. Bond, 1993. Evidence for general instability of past climate from a 250-kyr ice-core record. Nature 364, 218-220.

Diaz, H.F. and N.E. Graham, 1996, Recent changes in tropical freezing heights and the role of sea surface temperature. Nature 383, 152-155.

Doake, C.S.M. and D.G. Vaughan, 1991. Rapid disintegration of the Wordie Ice Shelf in response to atmospheric warming. Nature 350, 328.

Fahnestock, M., 1996. An ice shelf breakup. Science 271, 775-776.

Grootes, P.M., M. Stuiver, L.G. Thompson and E. Mosley-Thompson 1989. Oxygen isotope changes in tropical ice, Quelccaya, Peru. J. Geophysical Research 94D, 1187-1194.

Hastenrath, S. and P.D. Kruss, 1992. The dramatic retreat of Mount Kenya’s glaciers 1963-87: greenhouse forcing. Annals of Glaciology 16, 127-133.

Holdsworth, G. and E. Peake, 1985. Acid content of snow from a mid-troposphere sampling site on Mt. Logan, Yukon Territory, Canada. Annals of Glaciology 7, 153-160.

I.A.I., 1995. El Nino-Southern Oscillation and Interannual Climate Variability.  Report IAI/OES/5.DD.  The Inter-American Institute for Global Change Research, Washington D.C., 56pp.

Jouzel, J., N.I. Barkov, J.M. Barnola, C. Genthon, Y.S. Korotkevich, V.M. Kotlakov, M. Legrand, C. Lorius, J.P. Petit, V.N. Petrov, G. Raisbeck, D. Raynaud, C. Ritz and F. Yiou, 1989. Global changes over the last climatic cycle from the Vostok ice core record (Antarctica). Quaternary International 2, 15-24.

Koerner, R.M., 1977, Devon Island Ice Cap: core stratigraphy and paleoclimate. Science 196, 15-18.

Mann, M.E., J. Park, and R.S. Bradley, 1995. Global inter-decadal and century-scale climate oscillations during the last half millennium. Nature 378, 266-270.

Mayewski, P.A., L.D. Meeker, S. Whitlow, M.S. Twickler, M.C. Morrison, R.B. Alley, P. Bloomfield, and K. Taylor, 1993a. The atmosphere during the Younger Dryas. Science 261, 195-197.

Mayewski, P.A., G. Holdsworth, M. Spencer, S. Whitlow, M.S. Twickler, M.C. Morrison, K.K. Ferland, and L.D. Meeker, 1993b. Ice-core sulfate from three northern hemisphere sites: source and temperature forcing implications. Atmospheric Environment 27A, 2915-2919.

Mercer, J.H., 1978, West Antarctic ice sheet and CO2 greenhouse effect; a threat of disaster. Nature 271, p. 321-325.

Mitchell, J.F.B., S. Manabe, V. Meleshko and T.Tokioka, 1990. Equilibrium climate change - and its implications for the future.  In Climate Change: The IPCC Assessment.  J.T. Houghton, G.J. Jenkins and J.J. Ephraums (eds.).  Cambridge University Press, Cambridge, 131-172.

Monaghan, M.C. and G. Holdsworth, 1990. The origin of non-sea-salt sulphate in the Mount Logan ice core. Nature 343, p. 245-248.

Naruse, R., M. Aniya, P. Skvarca and G. Casassa, 1995. Recent variations of calving glaciers in Patagonia, South America, revealed by ground surveys, satellite-data analyses and numerical experiments. Annals of Glaciology 21, 297-303.

Naruse, R., P. Skvarca and Y. Takeuchi, 1997. Thinning and retreat of Glaciar Upsala, and an estimate of annual ablation changes in southern Patagonia. Annals of Glaciology 24, (in press).

Peel, D.A., 1993. Ice core evidence from the Antarctic Peninsula region. In Climate since A.D. 1500. R.S. Bradley and P.D. Jones (eds.). Routledge, London, 549-571.

Ribstein, P., E. Tiriau, B. Francou, and R. Saravia, 1995. Tropical climate and glacier hydrology: a case study in Bolivia. Journal of Hydrology 165, 221-234.

Rott, H., P. Skvarca, and T. Nagler, 1996. Rapid collapse of northern Larsen Ice Shelf, Antarctica. Science 271, p. 788-792.

Schubert, C., 1992. The glaciers of the Sierra Nevada de Mérida (Venezuela): a photographic comparison of recent deglaciation. Erdkunde 46, 58-64.

Skvarca, P. and R. Naruse, 1997. Dynamic behavior of Glaciar Perito Moreno, southern Patagonia. Annals of Glaciology 24, (in press).

Takeuchi, Y., R. Naruse and P. Skvarca, 1996. Annual air-temperature measurement and ablation estimate at Moreno Glacier, Patagonia. Bulletin of Glacier Research 14, 23-28.

Taylor, K.C., G.W. Lamorey, G.A. Doyle, R.B. Alley, P.M. Grootes, P.A. Mayewski, J.W.C. White, and L.K. Barlow, 1993. The 'flickering switch' of late Pleistocene climate change. Nature 361, 432-436.

Thompson, L.G., and E. Mosley-Thompson, 1987. Evidence of abrupt climatic change during the last 1500 years recorded in ice cores from the tropical Quelccaya Ice Cap, Peru. In: Abrupt Climatic Change, W.H. Berger and L.D. Labeyrie (eds.) Reidel, Dordrecht, 99-110.

Thompson, L.G., E. Mosley-Thompson and B.M. Arnao, 1984a. Major El Niño/Southern Oscillation events recorded in stratigraphy of the tropical Quelccaya Ice Cap. Science 226, 50-52.

Thompson, L.G., E. Mosley-Thompson and P.M. Grootes, M. Pourchet and S. Hastenrath, 1984b. Tropical glaciers: potential for ice core paleoclimatic reconstructions. J. Geophysical Research 89D, 4638-4646.

Thompson, L.G., E. Mosley-Thompson, J.F. Bolzan and B.R. Koci, 1985. A 1500 year record of tropical precipitation in ice cores from the Quelccaya Ice Cap, Peru. Science 229, 971-973.

Thompson, L.G., M. Davis, E. Mosley-Thompson and K. Liu, 1988. Pre-Incan agricultural activity recorded in dust layers in two tropical ice cores. Nature 336, 763-765.

Thompson, L.G., E. Mosley-Thompson, M.E. Davis, N. Lin, T. Yao, M. Dyurgerov and J. Dai, 1993. "Recent warming": ice core evidence from tropical ice cores, with emphasis on central Asia. Global and Planetary Change 7, 145-156.

Thompson, L.G., E. Mosley-Thompson, M.E. Davis, P-N. Lin, K.A. Henderson, J.Cole-Dai, J.F. Bolzan and K-B. Liu, 1995. Late Glacial Stage and Holocene tropical ice core records from Huascarán, Peru. Science 269, 46-50.

Warren, C.R., and D.E. Sugden, 1993. The Patagonian Icefields: A glaciological review. Arctic and Alpine Research 25, 316-331.

Yamada, T., 1987. Glaciological characteristics revealed by a 37.6 m deep core drilled at the accumulation area of San Rafael glacier, the northern Patagonia icefield. Bulletin of Glacier Research 4, 59-67.