Multiple linear regressions (MLRs), generalized additive models (GAMs), and artificial neural networks (ANNs) were compared as methods to forecast recruitment of Gulf of Alaska walleye pollock ( Theragra chalcogramma ). Each model, based on a conceptual model, was applied to a 41-year time series of recruitment, spawner biomass, and environmental covariates. A subset of the available time series, an in-sample data set consisting of 35 of the 41 data points, was used to fit an environment-dependent recruitment model. Influential covariates were identified through statistical variable selection methods to build the best explanatory recruitment model. An out-of-sample set of six data points was retained for model validation. We tested each model’s ability to forecast recruitment by applying them to an out-of-sample data set. For a more robust evaluation of forecast accuracy, models were tested with Monte Carlo resampling trials. The ANNs outperformed the other techniques during the model fitting process. For forecasting, the ANNs were not statistically different from MLRs or GAMs. The results indicated that more complex models tend to be more susceptible to an overparameterization problem. The procedures described in this study show promise for building and testing recruitment forecasting models for other fish species.

Measurements of nutrients, primary (14C) productivity and new (nitrate) productivity collected in the southeastern Bering Sea in July and August 2004 together with measurements from July 1981 are used to characterize the regional production system during summer and to analyze the major physical and nutrient factors controlling biological patterns. The patterns of surface productivity suggested by the two different summers were remarkably similar in the regions of data overlap and agree with recent compilations of summer conditions from remotely sensed data. Summer was characterized by low production in the middle and outer shelf associated with surface mixed layers that were stabilized by heat. Elevated productivity was restricted to regions associated with the surface expression of thermohaline fronts, as well as to tidally mixed regions adjacent to and between the Pribilof Islands. Based on Chl a and productivity measurements, the biological impact of the shelf-break front of the southeastern Bering Sea extends at least as far inshore as the Pribilof Islands. The existence of elevated summer productivity associated with a recently identified shelf-partition front between the Pribilof domain and the middle shelf domain east of St. Paul Island was confirmed, and its thermohaline nature resulting from cumulative intrusions of slope-derived water is described. Yearly estimates of new and primary productivity compiled for the various biophysical regimes range from >1800mmolNm−2yr−1 and >170gCm−2yr−1 at the shelf-break front to less than 700mmolNm−2yr−1 and 100gCm−2yr−1 in oceanic and coastal waters. We hypothesize that trace metals, rather than macronutrients or grazing, likely restrict the seaward extent of elevated production at the shelf-break front. Iron may be supplied to the front by isopycnal mixing from deeper shelf waters that were in contact with sediments. By extending elevated production throughout the summer, the thermohaline and tidal fronts enhance the coupling of primary production to epipelagic food webs for the large populations of fish, birds and mammals in the region. Changes in the intensity or location of these frontal systems may impact populations that forage in them from the nearby Pribilof Islands.

Many of the factors affecting recruitment in marine populations are still poorly understood. complicating the prediction of strong year classes. Despite numerous attempts, the complexity of the problem often seems beyond the capabilities of traditional statistical analysis paradigms. This study examines the utility of four statistical procedures to identify relationships between recruitment and the environment. Because we can never really know the parameters or underlying relationships of actual data, we chose to use simulated data with known properties and different levels of measurement error to test and compare the methods, especially their ability to forecast future recruitment states. Methods examined include traditional linear regression, non-linear regression, Generalized Additive Models (GAM), and Artificial Neural Networks (ANN). Each is compared according to its ability to recover known patterns and parameters from simulated data, as well as to accurately forecast future recruitment states. We also apply the methods to published Norwegian spring-spawning herring (Clupea harengus L.) spawner-recruit-environment data. Results were not consistently conclusive, but in general, flexible non-parametric methods such as GAMs and ANNs performed better than parametric approaches in both parameter estimation and forecasting. Even under controlled data simulation procedures, we saw evidence of spurious correlations. Models fit to the Norwegian spring-spawning herring data show the importance of sea temperature and spawning biomass. The North Atlantic Oscillation (NAO) did not appear to be an influential factor affecting herring recruitment. (c) 2005 International Council for the Exploration of the Sea. Published by Elsevier Ltd. All rights reserved.

The Pribilof Islands region is recognized to be a major nursery for age-0 walleye pollock (Theragra chalcogramma) in the eastern Bering Sea. However, the Pribilof Islands' importance for the recruitment of pollock in the Bering Sea has not been quantified. In this paper, we define the geographic extent of the Pribilof Islands region based on ocean circulation, location of major fronts, and trophic interactions. The abundance and distribution by oceanographic habitat (nearshore, frontal and offshore), of acoustically determined age-0 pollock are compared between the Pribilof Islands and the Inner Front of the eastern Bering Sea, for the summers of 1997-1999. Net samples collected during surveys in 1995-2002 are used to compare the abundance of age-0 pollock in the Pribilof Islands region to locations eastward along the Bering Sea shelf.We found densities of age-0 pollock near the Pribilof Islands to be generally higher than on the surrounding shelf, and higher or non-significantly different from the Inner Front. Through synthesis of our data and inferences of pollock growth and mortality, we estimate the Pribilof Islands region to have comprised 12.5% (1997), 53.7% (1998), and 24.7% (1999) of the total stock of age-0 pollock in the eastern Bering Sea. Since the Pribilof Islands region comprises only 14% of the total eastern Bering Sea area, evidence is given that a disproportion ally large fraction of the age-0 pollock population inhabits the Pribilof Islands nursery. (C) 2005 Elsevier B.V. All rights reserved.

This special issue provides the first comprehensive, interdisciplinary examination of the marine ecosystem of the eastern and central Aleutian Archipelago – an area that is susceptible to climate shifts, anthropogenic influences and ecosystem change. Relatively little focused research had been conducted in this region before the studies presented here. The earlier foundation of knowledge was largely from efforts of the National Marine Fisheries Service and the Outer Continental Shelf Environmental Assessment Program. Present investigations of the ecosystem of the Aleutian Islands were instigated by a marked decline in the western stock of Steller sea lions (Eumetopias jubatus) from more than 250 000 in the 1970s to less than 50 000 in the early 1990s. This decline resulted in the western stock of Steller sea lions being declared threatened in April 1990. The cause of their decline was not known, but the most likely mechanisms included: climate shifts, direct or indirect effects of fisheries, diseases, and top-down control through predation by killer whales. As part of the effort to evaluate these competing hypotheses and to understand why sea lion numbers continued to decline, Congress enacted legislation in 2001. That legislation funded investigations of the possible causes of the decline in abundance of the western stock of Steller sea lions that live from Kodiak Island in the Gulf of Alaska westward to and including the Aleutian Islands. The focus of this supplementary volume is the eastern and central Aleutian Archipelago (162–177°W), which is a critical habitat for the western stock of the Steller sea lion population and also where their decline has been the greatest. Many papers in this volume present results from just two cruises on the R/V Alpha Helix, one in 2001 and the second in 2002. These cruises were funded by the Steller Sea Lion Program, directly and through NOAA's Coastal Ocean Program. While the original focus of the volume was on Steller sea lions and the findings of these cruises, the volume soon extended beyond just sea lions and resulted in the first integrated ecosystem study of the waters surrounding the Aleutian Islands. Drs J. Schumacher and G. Kruse consented to contribute their time and expertise as guest editors. While both of them are knowledgeable of regional issues, they were not part of any Aleutian research team. Thus, they could be objective in judging the quality and content of manuscripts submitted for publication in this supplement. Rather than delivering the more typical editorial discussing the impact of this supplement on marine science, they elected to contribute a unique editorial on the importance of sustaining the ecosystem services of the Aleutian Archipelago. In their article, they call for development of an integrated ecosystem management plan that would involve the major agencies and stakeholders that are active in the region. This valuable addition thus recognizes and amplifies the important and timely theme of adopting an ecosystem approach to resource management. While there have been a number of isolated, historical studies of the marine ecosystem of the eastern and central Aleutian Archipelago, the results presented in this volume represent the first integrated ecosystem study of this productive region. In addition to observations on physical, chemical and lower trophic level characteristics of the region, papers are presented on unique, cold-water corals that have colonized many of the Aleutian Passes, on the paleoclimate of the North Pacific, and on studies of the fish, birds and mammals that inhabit this region. Major findings include a better understanding of the mechanisms that control flow through the passes and the magnitude of that flow; the importance of the medium-size passes (water depth between 130 and 200 m) in supplying nutrients to the Bering Sea euphotic zone; the importance of the small and medium-size passes in providing foraging for birds and mammals; recognition that the Aleutians may support the greatest abundance and diversity of cold-water corals in the world; and, that during the last millennia, climate variability has affected the ecosystem on many occasions. A single integrating factor was found: a partitioning of ecosystem characteristics occurs at Samalga Pass. Aleutian passes to the east of Samalga are more shelf-like in nature, and those to the west are more oceanic. Specifically, east of Samalga Pass there are mainly neritic zooplankton species, which support short-tailed shearwaters and high numbers of fish species. In contrast, west of Samalga there are mainly oceanic zooplankton, which provide food for northern fulmars and auklets, and fewer fish species. Euphausiids are important in diets of fish to the east of Samalga Pass, while copepods and myctophids are important in fish diets to the west. Similarly, pollock are the major prey of sea lions in the eastern part of the study area, and Atka mackerel fill that role to the west. There is also a marked division in the distribution of sperm whales, with none found east of the pass, and a greater diversity of cold-water corals and sponges found west of Samalga Pass. While this volume provides the best description and discussion of the eastern and central passes presently available, there remain many as yet unanswered questions. How will climate variability or regime shifts impact this narrow band of islands, especially any north–south shift in the predominant weather patterns of the region? Will any shift in the weather patterns result in a change in the transport of warm, nutrient-rich water through the passes, thus influencing the ecosystems of the Bering Sea? How will the large population of cold-water corals and sponges adjust to any change? The western stock of sea lions appears to have stabilized. Will they recover? It is our hope that the papers assembled in this volume will provide a solid basis on which to build future research and a source of new questions and hypotheses with which to structure future investigations. It is not enough to wait for further studies to provide answers to these questions. The existing body of knowledge needs to be incorporated into an overall plan for resource management that goes beyond the traditional interest of commercial fisheries. State and federal regulatory agencies must coordinate with other stakeholders (commercial and recreational fishers, subsistence users, oil and gas developers, shipping and tourism interests, etc.) to insure that the rich marine productivity and biodiversity of the Aleutian Archipelago is sustained. We acknowledge Blackwell Publishing and staff for helping us to bring this supplement to press. Dr David Checkley, Editor-in-Chief of Fisheries Oceanography, was supportive of this endeavor and helpful in bringing it to pass. We are indebted to our guest editors, Drs J. Schumacher (a.k.a. Two Crow) and G. Kruse, for their wisdom and many hours of labour. For research support, we are grateful to the NOAA Steller Sea Lion Program, the NOAA Coastal Ocean Program, Fisheries-Oceanography Coordinated Investigations, and the Arctic Section of the National Science Foundation. We thank the Cooperative Institute for Arctic Research at the University of Alaska Fairbanks, and the Joint Institute for Marine Observations at the University of California, San Diego, for their roles in facilitating this research. In particular, we acknowledge Dr Elizabeth Turner, NOAA Coastal Ocean Program, for partial funding for this supplement. Finally, many of the results presented here would not have been possible without the dedicated and skilled services of the crew and officers of the R/V Alpha Helix and the NOAA ship Miller Freeman. This research was sponsored by NOAA's Coastal Ocean Program and the Steller Sea Lion Research Program, and is contribution FOCI-L564 to Fisheries-Oceanography Coordinated Investigations and contribution 2845 to the Pacific Marine Environmental Laboratory.

Southeast Bering Sea Carrying Capacity (SEBSCC, 1996–2002) was a NOAA Coastal Ocean Program project that investigated the marine ecosystem of the southeastern Bering Sea. SEBSCC was co-managed by the University of Alaska Fairbanks, NOAA Alaska Fisheries Science Center, and NOAA Pacific Marine Environmental Laboratory. Project goals were to understand the changing physical environment and its relationship to the biota of the region, to relate that understanding to natural variations in year-class strength of walleye pollock (Theragra chalcogramma), and to improve the flow of ecosystem information to fishery managers. In addition to SEBSCC, the Inner Front study (1997–2000), supported by the National Science Foundation (Prolonged Production and Trophic Transfer to Predators: Processes at the Inner Front of the S.E. Bering Sea), was active in the southeastern Bering Sea from 1997 to 1999. The SEBSCC and Inner Front studies were complementary. SEBSCC focused on the middle and outer shelf. Inner Front worked the middle and inner shelf. Collaboration between investigators in the two programs was strong, and the joint results yielded a substantially increased understanding of the regional ecosystem. SEBSCC focused on four central scientific issues: (1) How does climate variability influence the marine ecosystem of the Bering Sea? (2) What determines the timing, amount, and fate of primary and secondary production? (3) How do oceanographic conditions on the shelf influence distributions of fish and other species? (4) What limits the growth of fish populations on the eastern Bering Sea shelf? Underlying these broad questions was a narrower focus on walleye pollock, particularly a desire to understand ecological factors that affect year-class strength and the ability to predict the potential of a year class at the earliest possible time. The Inner Front program focused on the role of the structural front between the well-mixed waters of the coastal domain and the two-layer system of the middle domain. Of special interest was the potential for prolonged post-spring-bloom production at the front and its role in supporting upper trophic level organisms such as juvenile pollock and seabirds. Of concern to both programs was the role of interannual and longer-term variability in marine climates and their effects on the function of sub-arctic marine ecosystems and their ability to support upper trophic level organisms.

Chris Baier, Nick Bond, Ric Brodeur, Troy Buckley, Lorenzo Ciannelli, Liz Conners, Chuck Fowler, Susan Henrichs, Jerry Hoff, Anne Hollowed, George Hunt, Jim Ianelli, Nancy Kachel, Carol Ladd, Allen Macklin, Lyn McNutt, Jeff Napp, Jim Overland, Sigrid Salo, Robert Schabetsberger, Jim Schumacher, Beth Sinclair, Alan Springer, Phyllis Stabeno, Al Tyler, Lucy Vlietstra, Muyin Wang, and Terry Whitledge

The goal of the Southeast Bering Sea Carrying Capacity (SEBSCC) program is to document the role of juvenile pollock in the eastern Bering Sea ecosystem. This includes examination of factors that affect their survival and development and testing of annual indices of pre-recruit (age-1) abundance (from the SEBSCC Concept Paper, July 1995). Within this framework, the purpose of the Indices Working Group (IWG, which was developed at the SEBSCC Principal Investigators meeting held in January 2001) is as follows. Based on the best understanding of ecosystem dynamics, identify potential single-or multi-parameter constructs or indices (e.g., wind mixing, time of spring bloom, etc.) that lead to development of survival indices for pollock in the egg, larval and young of the year life history stages. This information will provide input to the National Marine Fisheries Service (NMFS) stock assessment model and/or models of juvenile pollock for use by fisheries scientists at the Alaska Fisheries Science Center (AFSC)/NMFS. As noted by Napp et al (2000) pollock is the most abundant species harvested in the Bering Sea, accounting for > 65% of the total groundfish biomass and during the 1980s when their total biomass exceeded 20 million tones. The biomass trends of three major trophic guilds in the eastern Bering Sea (1979 to 1998: Schumacher et al., in press) show that while the total biomass of pollock in the 1990s is less than in the 1980s, they still dominate biomass in any of the guilds which include marine birds, mammals, other fishes and crabs. Walleye pollock is a nodal species in the food web (NRC Report, 1996) with juveniles being the It is natural that that pollock have been the focus of the Coastal Ocean Program's SEBSECC. The choice of developing a survival index for the early life history stages (eggs through young of the year) allows an early forecast of potential recruitment to the fishery and a metric that can be compared to existing time series of age-1 abundance (Figure 1). A switch model was developed for the eastern Bering Sea (Megrey et al., 1996), which identifies candidates for causing mortality by life history stage and the mortality variability of each stage indicating the stage contributing the most variability in recruitment to the fishery is set (Figure 2). In this model, transport and turbulence have their greatest impact on mortality of yoke-sack through feeding larval stages. It is the vertical structure of temperature, …

Two Bering Sea marine research programs collaborated during the final years of the 1990s to forge advances in understanding the southeastern Bering Sea pelagic ecosystem. Southeast Bering Sea Carrying Capacity, sponsored by NOAA Coastal Ocean Program, investigated processes on the middle and outer shelf and the continental slope. The Inner Front Program, sponsored by NSF, investigated processes of the inner domain and the front between the inner and middle domains. The purposes of these programs were to (1) increase understanding of the southeastern Bering Sea ecosystem, including the roles of juvenile walleye pollock, (2) investigate the hypothesis that elevated primary production at the inner front provides a summer-long energy source for the food web, and (3) develop and test annual indices of pre-recruit pollock abundance. The observations occurred during a period of unusually large variability in the marine climate, including a possible regime shift. Sea-ice cover ranged from near zero to one of the heaviest ice years in recent decades. Sea-surface temperatures reached record highs during summer 1997, whereas 1999 was noted for its low Bering Sea temperatures. Moreover, the first recorded observations of coccolithophore blooms on the shelf were realized in 1997, and these blooms now appear to be persistent. The programs’ results include an archive of physical and biological time series that emphasize large year-to-year regional variability, and an Oscillating Control Hypothesis that relates marine productivity to climate forcing. Further investigations are needed of the confluences of interannual and even intra-seasonal variability with low-frequency climate variability as potential producers of major, abrupt changes in the southeastern Bering Sea ecosystem.

NOAA's Fisheries Oceanography Coordinated Investigations (FOCI) contributes information to help forecast year-class strength of walleye pollock (Theragra chalcogramma) in the Gulf of Alaska. Quantitative estimates of recruitment are obtained from models of stock assessment and stock projection employing information supplied by FOCI. To generate its information, FOCI convenes specialists in marine biology, physical and fisheries oceanography, meteorology, and statistics to assemble and analyse relevant biological and physical time series with respect to recruitment and processes hypothesized to influence fish survival. Statistical methods encompass linear and nonlinear regression, stochastic simulation modelling, transfer function time series modelling, and tree-modelling, regression. The current database consists of 31 years of data, and analyses have identified factors that affect ocean stratification and circulation during spring and summer of the fish's birth year as being important to recruitment. A conceptual model of the recruitment process serves as the framework for a recruitment forecast scheme. A stochastic mathematical simulation model of the conceptual model produces similarities between simulated and observed recruitment time series. FOCI has successfully forecast recruitment observed over the past several years.

The 1991-93 El Niiio-Southern Oscdation (ENSO) event first appeared in the northern Gulf of Alaska in autumn 1991 with warm sea-surface temperatures. In winter 1992, there were pulses of increased sea level and anomalous circulation. El Niiio conditions persisted at least through summer 1993. The effects of this ENSO event on major groundfish species and Pacific herring in the northern Gulf of Alaska were examined and com- pared with the effects of previous ENSO events. There is little evidence that the 1991-93 or 1982-83 ENSO events affected landings of walleye pollock, Pacific cod, Pacific halibut, or arrowtooth flounder. Some changes in distribution of groundfish species were observed in 1993, but the effect was similar to changes observed in non-ENS0 warm years. In general, warm ocean condi- tions have a positive effect on recruitment of northern stocks, but ENSO events appear to have an inconsis- tent effect on year-class strength within species and among different species. For example, strong year classes of hal- ibut and arrowtooth flounder sometimes, but not always, coincide with ENSO events; ENSO events are associ- ated with moderate to weak year classes of cod and pollock. However, post-ENS0 warm years often are as- sociated with strong recruitment of many groundfish species. Major changes have occurred in the Gulf of Alaska ecosystem since 1977. The influence of the 1976 ENSO event in precipitating these changes and the role of the hequency or strength of subsequent El Nifio events is presently unknown. Herring and other stocks of small pelagic fishes may be more affected by ENSO events. In particular, decreased catches, recruitment, and weight- at-age of herring are sometimes associated with ENSO events. Furthermore, a variety of seabirds which feed mostly on pelagic forage fishes or the pelagic juvenile stages of groundfish suffered widespread mortalities and breeding failures in the Gulf of Alaska during the ENSO

The 1991-93 El Nino-Southern Oscillation (ENSO) event first appeared in the northern Gulf of Alaska in autumn 1991 with warm sea-surface temperatures. In winter 1992, there were pulses of increased sea level and anomalous circulation. El Nino conditions persisted at least through summer 1993. The effects of this ENSO event on major groundfish species and Pacific herring in the northern Gulf of Alaska were examined and compared with the effects of previous ENSO events. There is little evidence that the 1991-93 or 1982-83 ENSO events affected landings of walleye pollock, Pacific cod, Pacific halibut, or arrowtooth flounder. Some changes in distribution of groundfish species were observed in 1993, but the effect was similar to changes observed in non-ENSO warm years. In general, warm ocean conditions have a positive effect on recruitment of northern stocks, but ENSO events appear to have an inconsistent effect on year-class strength within species and among different species. For example, strong year classes of halibut and arrowtooth flounder sometimes, but not always, coincide with ENSO events; ENSO events are associated with moderate to weak year classes of cod and pollock. However, post-ENSO warm years often are associated with strong recruitment of many groundfish species. Major changes have occurred in the Gulf of Alaska ecosystem since 1977. The influence of the 1976 ENSO event in precipitating these changes and the role of the frequency or strength of subsequent El Nino events is presently unknown. Herring and other stocks of small pelagic fishes may be more affected by ENSO events. In particular, decreased catches, recruitment, and weight-at-age of herring are sometimes associated with ENSO events. Furthermore, a variety of seabirds which feed mostly on pelagic forage fishes or the pelagic juvenile stages of groundfish suffered widespread mortalities and breeding failures in the Gulf of Alaska during the ENSO years of 1983 and 1993. These effects on seabirds were also observed over a wider geographic range, from California to the western Bering Sea.

This study examines the possibility that wind rnixlng in Shelikof Strait, Gulf of Alaska, is a critical factor for larvae of walleye pollock Theragra chalcogramma. The abundances of walleye pollock larvae hatched on a given day and surviving through the early feeding stage were determined by in situ samplmg and otolith analysis for 1983 and 1985 to 1991. Periods of anomalously low or high larval survival were determined by comparing observed first-feeding date distributions of survivors sampled in late May surveys with expected first-feeding date distributions from a model utilizing information on spawning time and abundance, measured egg mortality, assumed larval mortality, and survey date. The cube of the wind speed represented daily estimates of mixing for the same years; wind speeds were determined from gridded sea-level pressure data using a geotriptic wind model. When the resulting daily dstributions of larval abundance and mixing were compared, 2 patterns emerged: (1) strong wind mixing events during the first-feeding period were associated with periods of lower than expected larval survival, and (2) periods of higher than expected larval survival were associated with calm wind periods often bracketed by strong mixing. The results indicate that over the 8 yr of observation strong mixing events during the first-feeding period were detrimental to survival of pollock larvae.

Surface winds measured from April 11, 1987, through August 16, 1987, at five locations off the southeastern coast of the Alaska Peninsula indicate the complexity of coastal wind structure as a function of proximity to a mountainous coast. Proxy winds were computed for the same five locations from digitized sea level pressure fields using a simple geotriptic wind model. Shoreward of the Rossby deformation radius, the friction parameters of the geotriptic model vary from their open-ocean values, slowing the winds and rotating them farther counterclockwise. These changing friction parameters account for the effect of the mountains on the surface wind distribution. Proxy winds and observed surface winds are coherent for periods longer than about 2 days and are better correlated farther seaward from the coast and for northeasterly alongshore winds and southeasterly onshore winds. Estimates of numbers of storms, regional wind stress, and regional vorticity determined from proxy winds are qualitatively accurate.

The effects of the upstream orography of the Alaska peninsula on the low-level flow in the coastal region are studied using observations from two NOAA P-3 research flights. The terrain in this region includes a low sill at Wide Bay (approximately 300 m high and 80 km wide), which is flanked by moderate terrain to the southwest (approximately 900 m high) and higher terrain to the northeast (approximately 1500 m high). For the case of 26 February 1987, a large Froude number (Fr approximately 1.6) characterized the incident flow. Boundary-layer wind speeds were approximately 30 m s-1 downstream of the gap at Wide Bay and the moderate terrain. The cross-terrain component of the wind above the boundary layer was 24 m s-1 upstream of the barrier and as large as 45 m s-1 approximately 70 km downstream of the barrier. Wind speeds were significantly less above and downstream of this wind maximum, as with a hydraulic jump. A prominent trough in sea level pressure was observed in the lee of the higher terrain; the largest 100-m wind speeds (approximately 34 m s-1) observed were near this trough. For the case of 3 March 1987, the incident flow over the Alaska peninsula was weak, and the Froude number was small (Fr approximately 0.4). In this situation, a low-level outflow (approximately 300 m high) with large wind speeds, cold air temperatures, and high sea level pressure was isolated to the region downstream from the gap at Wide Bay.

During February 1982 a NOAA research aircraft investigated a cold, low-level jet blowing from a gap between mountain ranges on the west side of Cook Inlet, Alaska. The jet blew 200 km southeastward across Cook Inlet between the Kenai Peninsula and the Kodiak archipelago, passing into the Gulf of Alaska where it merged with the large-scale marine wind field. Measurement commenced approximately 35 km downstream of the seaward end of the gap. The jet's internal boundary layer accelerated by 5% and grew 20% in depth for approximately 50 km; thereafter, wind speed and boundary-layer depth were nearly constant for the next 100 km. The strongest winds (> 20 m s-1 at a height of 80 m) were observed on the south side of the jet's thermal axis and 90 km downstream from the coast. Budget analyses reveal that the down-gradient acceleration within the jet was principally opposed by surface friction, and the cold air advection was balanced by a strong upward-directed sensible heat flux from Cook Inlet and entrainment of warmer air from aloft.

The thermal contrast between cold air over continental Alaska and relatively warm marine air over the Gulf of Alaska causes frequent, low-level, offshore-directed winds over the south-central Alaskan coast during the cold season. Coastal mountains affect these winds by inhibiting low-level mixing of continental and marine air masses near the coast, by providing channels that focus and accelerate drainage winds, and by exciting mountain-lee waves. Offshore-directed winds were observed twice with a research aircraft. The strongest winds were measured at the mouth of the Copper River and over and downwind of Resurrection Bay. The synoptic weather pattern and its orientation to local topographic features influenced wind magnitude. With the pressure gradient perpendicular to the coast, offshore-directed flow was light except at the Copper River and Resurrection Bay drainages. When the pressure gradient was aligned with the shore, regional surface winds were stronger and more uniform, although the Copper River and Resurrection Bay drainages wore still discernible. Evaluation of local force balances showed the largest ageostrophy at coastal locations downwind of bays and river valleys. Seaward from the coast, ageostrophic accelerations and cross-isobaric wind components were smaller, indicating a transition toward geotriptic equilibrium. This spatial adjustment pattern is consistent with the hypothesis that equilibrium is achieved within a distance similar to the regional Rossby radius of deformation.