Ironically, the chilly Arctic is one of the hot spots of global climate change: Temperatures there are increasing faster than anywhere else on Earth, coinciding with other rapid environmental changes across the region. Shifting conditions in the Arctic, remote as they may seem, will likely have substantial effects of interest to us humans living at lower latitudes, from altering fisheries and wildlife habitats to opening new transoceanic shipping routes and causing shorelines of thawing permafrost to crumble and encroach on coastal communities.
Projecting the future of Arctic climate remains difficult, however. This issue arises in part because of inconsistencies among results from climate modeling and analysis methods regarding potential links between Arctic warming and another element of Earth’s climate that dramatically affects humanity: severe weather in the midlatitudes, including storms and extreme temperatures. To help resolve some of these uncertainties and provide better data for weather and climate models, which should translate into more realistic forecasts and projections, we are leading a novel aircraft campaign to observe transformations of air masses along their way into and out of the Arctic.
We call this effort the HALO-(AC)³ campaign. HALO is short for the High Altitude and Long Range Research Aircraft, a modified Gulfstream G550 [Stevens et al., 2019], funded by Infrastructure Priority Program SPP 1294, and (AC)³ refers to a German collaborative project called “Arctic Amplification: Climate Relevant Atmospheric and Surface Processes, and Feedback Mechanisms.” HALO-(AC)³ involves not only HALO but also the Polar 5 and Polar 6 research aircraft (two modified Basler BT-67 planes).
Scheduled for March–April 2022, all three aircraft will follow air masses to and from the Arctic, a novel form of observation for such studies. The resulting measurements will help us check and improve the ability of models to reproduce transformation processes that air masses undergo as they travel along southward or northward trajectories (i.e., meridional transport), allowing scientists to more realistically reproduce linkages between the Arctic and the midlatitudes.
The Arctic Amplification Effect
Arctic climate is changing at an unprecedented pace and intensity [Jeffries et al., 2013]. Beginning a couple of decades ago, for example, scientists observed a remarkable increase in Arctic near-surface air temperatures, exceeding average global warming by at least a factor of 2 [Thoman et al., 2020].
Besides the exceptional warming, scientists are currently observing other dramatic Arctic climate changes, including the faster-than-expected decline of Arctic sea ice since about 1970 [Olonscheck et al., 2019]. Not all of this change in the Arctic is attributable to anthropogenic factors; for example, up to 50% of sea ice loss may be related to internal variability in the climate system. These changes are both consequences and drivers of a multitude of interlinked feedback mechanisms among climatic and Earth surface processes. The combined effects of these feedbacks are stronger in the Arctic than in other regions, leading to the phenomenon of Arctic amplification [Serreze and Barry, 2011].
A Network of Interconnected Systems
The underlying mechanisms of Arctic amplification involve myriad interactions in a complex physical system and are driven by unique conditions specific to the Arctic. Examples include pronounced temperature inversions (layers of warm air occurring above layers of cooler air) close to the ground, heterogeneous surface properties (e.g., open versus sea ice–covered ocean), a cold halocline ocean layer in which salinity levels change dramatically with depth, and persisting low-level clouds that often contain both liquid water droplets and ice crystals.
These Arctic-specific conditions may foster a number of local feedback mechanisms involving interactions among surface albedo, water vapor, clouds, lapse rate (the decrease in air temperature with increasing altitude), and Planck feedback processes (in which a warming Earth surface and near-surface air temperature emit more infrared radiation into space, counteracting the warming at the surface). Furthermore, remote feedbacks may link the Arctic and midlatitudes via meridional transport of heat, humidity, and momentum in the atmosphere and ocean [Cohen et al., 2014]. In this regard, large-scale atmospheric dynamical mechanisms related to Rossby waves (which are associated with rotating fluids like the atmosphere) could play a crucial role in Arctic amplification.
It has been suggested that increased “waviness” of the atmospheric jet stream (related to the enhanced magnitude of Rossby waves) is a response to the rapid Arctic warming [Francis and Vavrus, 2015]. However, other researchers contend that the causality likely goes in the opposite direction, with internal variability in the waviness of midlatitude circulation causing anomalous Arctic amplification [Blackport and Screen, 2020].
Either way, it has been established that Rossby waves cause meridional air mass transport into and out of the Arctic, and these air mass movements influence the energy budget of the Arctic atmosphere. This influence is especially pronounced during polar night in winter, when solar radiation, the major energy source driving the Arctic climate, is missing. Meridional transport includes northbound warm air intrusions (WAIs) as well as southbound cold air outbreaks (CAOs). Whereas WAIs inject warm, humid air into the Arctic, large-scale CAOs export cold, dry air out of the Arctic and may cause severe winter weather in the midlatitudes. Both phenomena thus play key roles in the Arctic climate system and influence its potential linkages to midlatitude weather [Pithan et al., 2018].
Pumping Warm, Moist Air into the Arctic
WAIs typically appear as filamentary, spatially localized phenomena, often occurring in pulses (Figure 2). Many WAIs enter the Arctic via the Atlantic in winter or early spring, and during these seasons, WAIs occur roughly once a week on average, taking up to 5 days to cross the Arctic [Pithan et al., 2018].
WAIs increase near-surface air temperatures because they contain water vapor that increases the proportion of terrestrial radiation that propagates downward through the atmosphere rather than escaping into space. About a third of all water vapor transport events into the Arctic are related to WAIs. Despite this relative rarity, WAI-related water vapor transport events contribute roughly two thirds of the total water vapor mass transported into the Arctic in winter.
WAIs also increase wintertime cloudiness in the Arctic, further enhancing the amount of downward terrestrial radiation. These water vapor and cloud effects are more efficient in heating near-surface Arctic air than direct south–north heat transport caused by atmospheric flow (advection). Overall warming due to WAIs is estimated to increase near-surface air temperatures in winter by about 5 K. This warming may trigger an earlier onset of sea ice melting and more intense development of melt ponds, both of which decrease surface albedo and lower the ability of sea ice to reflect solar radiation.
Altogether, WAIs transport warm air and water vapor into the Arctic and trigger the development of clouds that tend to be long-lived, typically lasting for days. Both processes heat the surface and may contribute to Arctic amplification.
Cold, Dry Air Spills Southward
CAOs can initiate polar low-pressure systems (polar lows) over the Arctic Ocean and cool midlatitude continents and subpolar oceans, potentially leading to extreme midlatitude weather. Arctic CAOs occur most frequently in winter, with about 30 such wintertime events annually. The southward moving air masses during CAOs are modified by large differences in surface air temperatures over sea ice versus over open ocean water, which can exceed 30 K [Pithan et al., 2018]. Sea ice−covered and open ocean areas are separated by the transitional marginal sea ice zone, and properties of CAO air masses are also influenced by surface temperature differences in this zone in areas called ice leads and polynyas—transient and stable gaps in the sea ice, respectively. The extreme temperature gradients the air masses endure cause large heat releases from the ice-free ocean to the much colder air. The amount of heat lost over leads and polynyas depends on the surface area of water exposed, but the formation of leads and polynyas is not well represented in models.
CAOs may account for 60%–80% of the heat transferred from the ocean to the atmosphere in the Arctic. These transfers cause rapid warming and moistening of air masses moving southward, which initiates atmospheric instability that may affect ocean mixing and sea ice formation. Clouds evolve during marine CAOs, first appearing as cloud streets (roll circulation) and later evolving into open cells. These processes are often insufficiently represented in models; improving their realistic representation is important for enhancing the predictive performance of weather and climate models in the Arctic and midlatitudes [Pithan et al., 2018].
Where Models Fall Short
General circulation models (GCMs) used to simulate future climate provide different conclusions about the influence of Arctic warming on midlatitude weather [Cohen et al., 2020]. For example, Coupled Model Intercomparison Project Phase 5 (CMIP5) models systematically underpredict the moisture flux from the Atlantic sector into the Arctic compared with the European Reanalysis (ERA) Interim.
Furthermore, GCMs are not good at simulating cloud evolution and boundary layer development during air mass transformation processes related to WAIs and CAOs that can trigger additional local feedback mechanisms. For example, most climate models inadequately modulate clouds in general and especially misrepresent mixed-phase clouds (those containing both supercooled liquid water droplets and ice crystals), causing climate model biases with respect to Arctic wintertime temperature inversions.
Regional models, in particular, often fail to realistically represent cloud effects in the Arctic. Most tend to exaggerate the degree of ice formation in clouds, which may result from model parameterizations that overestimate ice crystal formation in mixed-phase clouds. However, there are also models that overestimate low-level, liquid-water-containing clouds in the Arctic. The inconsistencies among models indicate uncertainties about underlying causes in the representation of radiative and turbulent processes, which are influenced by ice formation.
To resolve potential reasons for discrepancies among and within individual models, we require detailed studies of the models’ abilities to reproduce key processes, such as cloud evolution during the transformation of the air masses along meridional pathways [Wendisch et al., 2019].
Reconciling Observations with Model Outputs
To date, observations of air mass transformations in the Arctic have mostly been conducted from fixed local positions (Eulerian point of view). Few aircraft-based samplings of changes in air mass properties during transport exist, and these observations have covered only a limited region. The Eulerian approach does not permit the required observations of air mass–transforming processes.
For the HALO-(AC)3 campaign, we proposed a quasi-Lagrangian approach [Pithan et al., 2018], in which HALO and the other aircraft follow air masses to and from the Arctic to observe them directly during WAIs and CAOs. (We intend to focus on WAIs, which have been observed much less frequently than CAOs.) This sampling strategy is called quasi-Lagrangian because the HALO aircraft is naturally much faster than the relatively slow air masses, so the aircraft cannot move in total synchronicity with an air parcel; rather, it meets and samples the air mass along its trajectory as often as possible during dedicated flight patterns.
The HALO-(AC)³ mission has two main objectives. The first is to use HALO, operated by the German Aerospace Center, to perform quasi-Lagrangian observations of air mass transformation processes during meridional transports within WAIs and marine CAOs, an approach that has not been tried before in the Arctic. The second is to test the ability of numerical atmospheric models to reproduce the measurements taken from the aircraft. The tested models can then be applied to investigate linkages between Arctic amplification and midlatitude weather.
Figure 3 illustrates the intended observational strategy to achieve the first objective. We will observe the transformation processes of air masses as they progress over open water, the marginal sea ice zone, and the sea ice of the Arctic Ocean during northbound WAIs and vice versa for southbound marine CAOs.
HALO is the only research aircraft available in Germany that has the required endurance—it can stay aloft for up to 10,000 kilometers or up to 9 hours at a time—for the planned quasi-Lagrangian observations. It can also carry state-of-the-art meteorological and remote sensing instruments (up to 3 metric tons) high enough (up to 15 kilometers) to observe the complete vertical tropospheric air mass column, including meteorological quantities, turbulence and radiation parameters, water vapor, aerosol particles, and clouds.
The remote sensing payload HALO will carry in the upcoming mission has matured over the course of several past campaigns [Stevens et el., 2019]. The instrumentation includes a 26-channel microwave radiometer, a Ka band Doppler radar, and aerosol and differential absorption lidar instruments for measuring water vapor. HALO also has spectral and broadband solar and thermal-infrared radiation sensors (upward and downward looking) and imaging camera spectrometers that detect in the solar and thermal infrared spectral ranges, as well as numerous dropsondes (instruments dropped from the plane that take data as they fall to Earth).
The low-flying Polar 5 and Polar 6, belonging to the Alfred Wegener Institute (AWI), will operate in tandem with HALO. Polar 5 will provide active and passive remote sensing measurements, and Polar 6 will make in situ measurements of clouds, aerosol particles, and radiation. The AWI aircraft will enable us to probe radiative and turbulent energy fluxes and smaller-scale processes in the lower troposphere (below 3–5 kilometers altitude) and to observe surface properties. This strategy corresponds to observations performed during the (AC)3’s 2017 campaign called Arctic Cloud Observations Using airborne measurements during polar Day (ACLOUD) [Wendisch et al., 2019].
Flight transects are intended to follow an air parcel for at least three consecutive days and to revisit the parcel every day. We will track thermodynamic, cloud, and aerosol characteristics along and perpendicular to the trajectories. Dropsondes released during circular flight patterns will record large-scale vertical air motion.
We will pursue the project’s second objective, testing the ability of numerical atmospheric models to reproduce the aircraft measurements, by building case studies. Each case will focus on a specific WAI or CAO, for which we will collect measurements conducted over sea ice, the marginal ice zone, and open ocean. We will use these data to investigate the effects of distance to the ice edge, ice concentration, lead fraction, and ice topography. The HALO-(AC)³ mission will provide unique data to evaluate the output of numerical atmospheric models covering a wide range of scales, both spatial (from single air columns to Arctic-wide) and temporal (from instantaneous to several days).
Models will range from large-eddy simulations that, for example, follow the motion of an air column within an air mass sampled by HALO to mesoscale (from 5 to hundreds of kilometers) atmospheric models. Using the measurements from HALO-(AC)³, we will perform sensitivity studies to isolate important individual processes, such as cloud evolution during WAIs, and to test assumptions used to represent turbulence and clouds in climate models.
Involving modeling groups in the flight planning will be crucial to the mission because they can provide predicted trajectories of the air parcels to be sampled. We hope to return the favor by providing the modeling groups with data that enable unique insights into the interacting processes that drive the Arctic atmosphere and climate. The knowledge we gain from these improved models will be essential in understanding Arctic amplification and linkages between the Arctic and midlatitude climate, as well as in predicting and responding to the continuing changes in Arctic climate that will affect humanity in the years ahead.
Acknowledgments
We thank Johannes Quaas and Karolin Block for providing Figure 1, André Ehrlich for designing Figure 3, and Andreas Macke for the photograph at the beginning of this article. We gratefully acknowledge funding by the Deutsche Forschungsgemeinschaft (DFG) project 268020496–TRR 172, within the Transregional Collaborative Research Center “Arctic Amplification: Climate Relevant Atmospheric and Surface Processes, and Feedback Mechanisms (AC)3.” This work was also supported by DFG Infrastructure Priority Program (Schwerpunktprogramm) SPP 1294.