Through ARISE, NASA acquired unique aircraft data on clouds, atmospheric radiation and sea ice properties during the critical period between the sea ice minimum in late summer and autumn and the commencement of refreezing.
Arctic sea ice decline is one of the most profound manifestations of contemporary climate change, and the loss has been accelerating in recent years as seen by regular extreme September minima and lengthening of the melt season by 5 days decade−1 (Stroeve et al. 2012, 2014). This overall decline, combined with a shift toward entirely seasonal ice (Perovich and Polashenski 2012), implies the action of numerous feedbacks involving thinner and darker ice, changing cloud cover, and increasing energy input to the upper water column. Radiation feedbacks are a necessary mechanism to drive this decline (Perovich et al. 2008), although anomalous winds and preconditioning also play a major role in both trends and variability (Zhang et al. 2008). At the same time, it is expected that this large-scale decrease in Arctic sea ice will drive circulation anomalies throughout the troposphere (Cassano et al. 2014). There is a need to diagnose these changes empirically, and to validate climate model simulations, on a pan-Arctic basis.
Ultimately, this need is most satisfactorily addressed with well-characterized satellite remote sensing data. Several sensors from the National Aeronautics and Space Administration (NASA)’s Terra and Aqua spacecraft and A-Train constellation (https://atrain.gsfc.nasa.gov/) have provided observations of key components of the Arctic climate system for more than a decade, including atmospheric structure, cloud optical properties, and sea ice concentration (sea ice being available in the passive microwave satellite record going back to 1979). Concurrently, the Cloud and the Earth’s Radiant Energy System (CERES) sensors, and their predecessors from the Earth Radiation Budget Experiment (ERBE), retrieve the net shortwave and longwave fluxes that reveal the combined action of the radiative and dynamical feedbacks involving Arctic sea ice. Hartmann and Ceppi (2014) use CERES data to show that every 106 km2 decrease in September Arctic sea ice in recent years corresponds to an annual-mean increase in absorbed shortwave radiation of 2.5 W m−2 between 75° and 90°N. Further progress in our understanding of the whole Arctic climate system requires understanding how the individual components of the Arctic ocean–atmosphere system manifest in the CERES-measured fluxes and how well they are retrieved by other satellite sensors.
In addition, high-quality spectral and broadband radiometric data from above sea ice, and below, within, and above Arctic stratiform clouds, can provide a valuable resource for testing the overall effectiveness of parameterizations for cloud and sea ice evolution in climate models. For example, if a regional model is initialized with the meteorological conditions pertaining to a given flight mission, then the simulated energy fluxes at the surface and below, within, and above cloud can be compared with the data to note where agreement or discrepancies occur. If general model–data agreement appears in the microphysics, for example, then discrepancies in measured irradiance may be related to the radiative transfer parameterization (e.g., three-dimensional effects vs a plane-parallel model). Comparison of Arctic surface radiation measurements with climate model simulations has proven valuable (Tjernström et al. 2008); however, to date most Arctic aircraft studies related to climate model parameterizations have concentrated on cloud microphysics (e.g., Fridlind et al. 2007, 2012). Here we describe a unique aircraft campaign focused on cloud properties and radiative effects that can benefit both the remote sensing and climate modeling approaches to the study of Arctic change.
EXPERIMENT DESIGN AND EXECUTION.
One remarkable aspect of the Arctic Radiation-IceBridge Sea and Ice Experiment (ARISE) is the short timeline from experiment conception to successful execution in September 2014. NASA funding became available in March of 2014 to supplement Operation IceBridge (OIB) with sea ice observations during the September transition in the Beaufort–Chukchi Seas, and a C-130 aircraft (N439NA) was also available that was capable of carrying advanced instrumentation for cloud and atmospheric energy budget observations during a time frame that is relatively undersampled in the high Arctic compared with spring and midsummer. OIB is an ongoing airborne science campaign to characterize sea ice, glaciers, and ice sheets in unprecedented detail while bridging the gap in polar observations between NASA’s Ice, Cloud, and Land Elevation Satellite (ICESat) missions. The sea ice, radiation, cloud microprobe, and meteorological instruments are listed in Table 1, and their aircraft installation is depicted in Fig. 1. Because of the unusually short planning timeline, much of the instrument selection was based on proven track records and uncomplicated installation in the C-130. Nevertheless, the instrument suite was comprehensive and advanced, yielding a timely dataset, preliminary results of which are presented here.
Parameters measured from the C-130 (NASA 439) during ARISE. NRL = Naval Research Laboratory. NSERC = Natural Sciences and Engineering Research Council of Canada. GSFC = Goddard Space Flight Center. LaRC = Langley Research Center.
The NASA C-130 research aircraft as configured for ARISE, showing the location of each instrument described in Table 1.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
While NASA satellites are making routine observations, an accurate interpretation of the data required to track Arctic climate change can be difficult. Uncertainties in atmospheric temperature and humidity, heterogeneity in surface conditions (including sea ice properties), and difficulties detecting and characterizing clouds over sea ice all contribute to the uncertainty associated with the CERES-derived irradiances, which is currently larger over sea ice than any other scene type (Su et al. 2015b). Thus, the evaluation of CERES top-of-atmosphere (TOA) and surface (SFC) radiative fluxes over the Arctic with data from the C-130 payload is a unique and important ARISE scientific objective. A number of ARISE flight plans were designed specifically to accomplish this objective over a wide range of conditions. Other flight plans were designed to characterize the composition of low-level clouds and their radiative effects over various sea ice conditions and to support OIB with sea and land ice characterizations. Recent work has shown that heterogeneity and small-scale interactions are important to consider, particularly in leads and over open water adjacent to sea ice (Vihma et al. 2014). The high time resolution of both the radiometric suite and surface remote sensors provides direct observation of heterogeneity.
ARISE was based at Eielson Air Force Base (AFB) near Fairbanks, Alaska. Weather prediction and regional modeling resources were used on-site for flight planning. Aircraft mission planning fell into three major categories: 1) CERES collocation and validation, 2) sea ice observation, and 3) cloud sampling. The missions that were accomplished are detailed in Table 2, and the associated flight tracks are illustrated in Fig. 2. Figure 3, obtained from the nadir and forward-looking cameras, shows examples of the wide variety of sea ice conditions sampled during ARISE, including thick multiyear ice, a wide range of broken and scattered ice conditions, melt ponds, and frazil and black ice upon refreezing.
ARISE mission summary: select satellite overpass times (A: Aqua, C: Cryosat-2, T: Terra, S: Suomi National Polar-Orbiting Partnership), dominant surface type, and flight description. KWAL = Wallops Flight Facility, Wallops Island, VA. KTCM = McChord AFB, Tacoma, WA. PAEI = Eielson AFB. SCT = scattered. BKN = broken.
Map of the ARISE mission flight tracks as described in Table 2.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
Surface conditions during the 2014 sea ice transition period over the Beaufort Sea photographed from the ARISE aircraft using the nadir camera. (a) 5 Sep, 80°N, 140°W; (b) 10 Sep, 77°N, 128°W; (c) 17 Sep, 74°N, 152°W; (d) 16 Sep, 77°N, 143°W; (e) 16 Sep, 76°N, 141°W; and (f) 24 Sep, 73°N, 129°W.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
The dates for the CERES experiments were fixed in advance, based on the known intersection of several satellite overpasses sufficiently within the range of the aircraft to allow for extensive gridbox flight patterns over the Beaufort Sea. Outside of those dates, sea ice and cloud radiation sampling missions were organized in near–real time based on the comprehensive weather data and forecasting available in the field. There was some advance planning given to within-cloud stacked transects, but due to the dynamic nature of the cloud cover, the cloud radiation missions more often adapted to the conditions on the spot. On these occasions, satellite meteorology observations and updated forecasts were transmitted to the aircraft en route to the Beaufort Sea, to help vector the mission to the most interesting scenes.
METEOROLOGICAL CONDITIONS.
Supporting weather forecasts for the ARISE flights were conducted with the NASA Goddard Earth Observing System Model, version 5 (GEOS-5; Molod et al. 2015), and Polar Weather Research and Forecasting (WRF) Model, version 3.5.1 (http://polarmet.osu.edu/PWRF/; Hines et al. 2015). Output fields from the forecasts are used here along with atmospheric reanalyses to represent synoptic meteorological conditions during the field program. Meteorology during ARISE may be categorized by two distinct regimes. During the first seven flights over the Arctic Ocean (4–11 September), the meteorological state was dominated by a surface high pressure over the southern Chukchi and/or Beaufort Seas. Figure 4 shows a composite set of 21-h Polar WRF forecasts valid at 1300 Alaska daylight time (AKDT), roughly at the midtimes of the C-130 flights. This resulted in northeasterly low-level flow over the Arctic coast and northern and central Alaska. There was considerable low-level cloudiness over the southern Beaufort Sea, consistent with the seasonal climatology (e.g., Intrieri et al. 2002). However midlevel and precipitating clouds were not extensive. Temperatures over central Alaska were mild with limited cloud cover—as indicated by the GEOS-5 cloud fraction (Fig. 5a), providing excellent flying weather.
Average sea level pressure (contours, interval 2 hPa) and 2-m temperature (°C; shaded) from Polar WRF 21-h forecasts (1300 AKDT) for (a) the first seven flights (4–11 Sep 2014) and (b) the next seven flights (13–21 Sep 2014).
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
Total cloud fraction from NASA’s Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2), for 2200 UTC, averaged for (a) 4–11 Sep and (b) 13–21 Sep 2014.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
A key synoptic shift occurred near 13 September that accompanied a northward advance and deepening of low pressure over Bristol Bay. Surface pressures fell over Alaska and the southern Beaufort Sea. During this second regime of 13–21 September, the region of surface high pressure was now located several hundred kilometers farther north over the Arctic Ocean (Fig. 4b). This resulted in east-northeasterly low-level flow over the flight target regions of the Arctic Ocean originating from a cold source region over sea ice. Simulated surface temperatures over the sea ice suggest surface freezing and thickening of the ice pack, consistent with reports from the C-130 staff (Fig. 4b). A weak time-averaged minimum pressure was located over the northwest corner of Alaska, as a series of weak mesoscale lows propagated eastward through the region. This is consistent with increased cloud cover over the North Slope of Alaska and the southern Beaufort Sea (Fig. 5b). Increased cloud cover and some light precipitation occurred in central Alaska during the second regime, and daily average temperatures dropped from near 15°C at Eielson on 13 September to 5°C on 21 September. During the later stages of this regime, dense fog occasionally appeared in the morning over central Alaska, limiting the C-130 flights from Eielson. Time series of Polar WRF low-level temperature over open ocean and sea ice in the Beaufort Sea indicate fluctuations on mesoscale and fast synoptic time scales between cold periods of strong low-level static instability and warmer periods of near-neutral low-level static stability (Fig. 6). Low-level temperatures were several degrees colder over sea ice than over open water. Moreover, the Polar WRF simulations show that during the ARISE field program faster net seasonal cooling occurred over sea ice than over open water.
Time series of skin temperature (solid lines) and 2-m temperature (dashed) for Polar WRF grid points at 73°N, 133°W (blue) and 73°N, 150°W (green). Temperatures are 21-h forecasts valid at 1300 AKDT, near the center times of ARISE flights.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
The Polar Meteorology Group at The Ohio State University has done extensive Arctic testing of Polar WRF, including in the northern Alaska and Beaufort Sea regions. Specific to the ARISE campaign, we compared a Polar WRF, version 3.6, run against near-surface observations from Barrow, Nome, Prudhoe, and Red Dog in Alaska, and buoys in the Chukchi Sea. Polar WRF was run on a 283 × 312 cell grid with 70 vertical levels and 8-km horizontal resolution. Table 3 shows that the model reasonably produces the near-surface air temperature, wind speed, wind direction, and surface pressure during September 2014. The multiday sea level pressure averages for regime 1 and regime 2, shown by Figs. 4a and 4b, respectively, are highly consistent with the summer and fall seasonal low-level wind climatologies near northern Alaska as shown by Figs. 3c and 3d in Zhang et al. (2016), respectively. Early analysis of the Polar WRF simulations suggest that ARISE meteorology during September 2014 yielded less low cloud liquid water and more cloud ice than during the August–September 2008 Arctic Summer Cloud Ocean Study (ASCOS; Tjernström et al. 2012).
Demonstration of monthly mean Polar WRF (<PWRF>), version 3.6, simulation agreement with Alaska and Chukchi Sea monthly mean observations (<Obs>) of near-surface temperature (°C), wind speed (m s−1), wind direction (°), and mean sea level pressure (MSLP, hPa), for Sep 2014. The surface observation stations are Prudhoe Bay (70.40°N, 148.53°W), Nome (64.50°N, 165.43°W), Klondike buoy (70.87°N, 168.25°W), Red Dog Dock (67.58°N, 164.07°W), Burger buoy (71.50°N, 164.13°W), and Barrow (71.29°N, 156.79°W).
PRELIMINARY RESULTS.
CERES.
CERES is a key component of the Earth Observing System (EOS) and Suomi National Polar-Orbiting Partnership (SNPP) observatory. During ARISE, four CERES instrument flight models (FM) were fully functional on the EOS Terra (FM1 and FM2), Aqua (FM3), and the SNPP (FM5) satellites. The CERES program strives for consistent instrument performance, calibration, and data products across satellite platforms to the extent possible. CERES products provide the most accurate spatially complete depiction of radiant energy exchanges in the Arctic. However, the uncertainty associated with the CERES-derived irradiances is currently larger over sea ice than any other scene type (Su et al. 2015b). The CERES Science Team provides instantaneous satellite footprint (level 2) and the hourly gridded mean (level 3) TOA and surface irradiance data products. ARISE observations provide an opportunity to evaluate irradiances for both of these products over the Arctic. Two CERES objectives are 1) to evaluate the level 2 CERES-derived top-of-atmosphere irradiance over areas with different sea ice conditions and 2) to evaluate hourly gridded mean irradiances in the level 3 CERES radiative flux data products.
The CERES instrument measures reflected and emitted shortwave (SW; 0.2–5 µm) and longwave (LW; 5–50 µm) radiances at a footprint size of ∼20 × 20 km at nadir. Loeb et al. (2012) demonstrate excellent stability of the CERES instrument to better than 0.3 W m−2 decade−1 and an absolute accuracy (2σ) of the CERES TOA fluxes of 2% in the SW and 1% in the LW (Loeb et al. 2009). After properly accounting for the spectral response of the radiometric filters (Loeb et al. 2001), the CERES radiances are converted to irradiances using angular distribution models (ADMs; Su et al. 2015a; Loeb et al. 2005). An ADM is a set of anisotropic factors that relates the radiance measured at a certain viewing geometry to a radiant flux. The anisotropy of the radiation field varies significantly under different surface types and cloud conditions. Thus, ADMs vary with scene type, especially for the shortwave, and accurate scene type identification is critical. The scene properties of each footprint are determined using a combination of satellite imager–derived cloud and surface properties (Minnis et al. 2011) and microwave-derived sea ice information. Temperature and humidity profiles required for the cloud retrievals are obtained from the NASA Global Modeling and Assimilation Office (GMAO) data assimilations system (Rienecker et al. 2008). Scene types in the Arctic are complex due to widely variable surface (e.g., Fig. 3) and cloud conditions.
To better evaluate the ADM performance and associated uncertainties in the instantaneous fluxes, one of the two CERES instruments on the Terra satellite—FM2—was placed in programmable azimuthal plane (PAP) scan mode during the ARISE campaign. The PAP mode was set to rotate FM2 for continuous targeting of a specific area as Terra passed over the region. This mode significantly increases the CERES sampling density and provides irradiance estimates over a wider range of viewing geometries in the area of interest. The other CERES instrument on Terra—FM1—was set to scan in the nominal cross-track direction. The difference in the spatial and angular sampling patterns for the FM1 and FM2 instruments is illustrated in Fig. 7. FM1 samples the broader area with a narrower viewing geometry, while FM2 samples over a more limited area but with a wider range of viewing geometries. This combination of coincident information from the PAP and cross-track scan modes, along with the aircraft measurements, provides a unique capability to test the CERES ADMs and thus evaluate the uncertainties associated with CERES level 2 TOA data products.
(a) TOA SW irradiance derived from the CERES FM1 instrument (operated in the cross-track mode) on Terra. The orange box indicates the area where the TOA gridbox experiment took place on 11 Sep 2014. (b) As in (a), but for the FM2 instrument on Terra that was operated in a programmable azimuthal-plane mode. (c) Viewing zenith and relative azimuth angles of CERES FM1 (red) and FM2 (blue) measurements inside the orange grid box in (a) and (b).
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
Collocated aircraft measurements with level 2 satellite observations have been previously used to evaluate instantaneous irradiances and retrievals from satellite instruments. However, these occur only over a short time window for a given satellite overpass, leading to a small sample size and significant noise in the comparisons. Even under a best-case scenario, where instantaneous satellite-derived irradiances are found to agree with aircraft measurements, the corresponding uncertainty for hourly 1° × 1° gridded radiant fluxes is not clear. Thus, the direct evaluation of level 3 TOA and surface irradiances is a major goal and a unique concept of the ARISE mission.
To create the level 3 data products, the level 2 CERES fluxes are aggregated to construct hourly 1° × 1° gridded mean TOA radiant fluxes (Doelling et al. 2013). The CERES Synoptic (SYN) level 3 data (CERES level 3) also contain hourly 1° × 1° gridbox-mean surface irradiances (Rutan et al. 2015). CERES level 3 atmospheric and surface irradiances are computed hourly. Surface radiant fluxes are evaluated using radiant flux measurements at surface sites (Rutan et al. 2015; Kato et al. 2013). Uncertainty in level 3 surface radiant fluxes is described in Kato et al. (2013). Over the Arctic Ocean, conventional observations of the surface and atmosphere are scarce and there are few opportunities to evaluate irradiances. Furthermore, the characterization of cloud and atmospheric conditions required for CERES irradiance computations is more uncertain over the Arctic than over other regions of the world. Thus, larger errors in CERES surface irradiances are also likely. ARISE observations enable an evaluation of CERES input datasets and the subsequent TOA and surface level 3 irradiances, which are extensively used in model evaluation (e.g., Pincus et al. 2008; Wang and Su 2013; Itterly and Taylor 2014; English et al. 2014).
To acquire the necessary data, the NASA C-130 flew “lawn mower” patterns (Fig. 2) over ∼200 km × 100 km or ∼100 km × 100 km grid boxes at a nearly constant altitude, either ∼6 km (TOA experiment) or near the surface (surface experiment), for 2–3 h. TOA experiment flight paths consisted of five legs of 200-km length, spaced 20 km apart. The surface flight paths consisted of seven 100-km-length legs, spaced 15 km apart. The flight paths corresponding to the TOA experiments are shown in Figs. 8a–c. TOA and surface experiments were conducted in pairs over a particular region, separated by 2 days. This pairing strategy allowed ARISE to capture aircraft measurements of TOA and surface irradiances along with other data over similar surface conditions, and with the most optimal coincidence with CERES and other satellite overpasses.
(a)–(c) Sea ice cover derived from Advanced Microwave Scanning Radiometer 2 (AMSR2) [Arctic Radiation and Turbulence Interaction Study (ARTIST) sea ice (ASI) algorithm] with NASA C-130 flight-track overlays on 3 days when TOA gridbox experiments were conducted. The CALIPSO ground track is indicated by the dashed red lines. The P and Q markers in (c) correspond to the P and Q points, respectively, in Fig. 9. (d)–(f) Distribution of CERES-derived LW and SW irradiances over the grid box encompassed by the orange solid lines shown in (a)–(c), respectively.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
One advantage of the Arctic compared to lower-latitude areas is the high frequency of polar-orbiting satellite overpasses that occur over a given region since the satellite orbits spatially converge. For ARISE, three “gridbox” locations were selected based upon the expected sea ice conditions and the most coincident satellite overpass times for the following spacecraft: Terra, Aqua, SNPP, Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO), and CloudSat. One flight leg of the lawn-mower pattern was always aligned with the CALIPSO/CloudSat ground track (Fig. 7, dashed red line). These active sensor observations, collocated with the aircraft data, provide detailed vertical profiles of clouds (Fig. 9) that are important to the evaluation of CERES irradiances, Moderate Resolution Imaging Spectroradiometer (MODIS) cloud retrievals, and the attribution of irradiance errors. For example, the MODIS cloud-top heights shown in Fig. 9d are retrieved with a single-layer assumption, which leads to underestimates when compared to CloudSat/CALIPSO retrievals in multilayered conditions. The MODIS cloud optical properties are also more uncertain over snow and ice for thinner clouds. Kato et al. (2011) demonstrate improvements in surface radiation budget estimates over polar regions when combining cloud properties from CALIPSO and CloudSat with MODIS data. More detailed analyses to determine how MODIS cloud retrieval errors contribute to the surface irradiance uncertainties, particularly when active sensor data are not available, remain as future work. Multilayer retrieval methods (e.g., demonstrated later in Fig. 14) and other improvements in MODIS cloud retrievals are being developed and evaluated with ARISE and A-train data.
Cloud-layer mask derived on 15 Sep 2014 from (a) CALIPSO vertical feature mask (VFM), (b) CloudSat 2B-CLDCLASS, (c) clouds detected by both CALIPSO and CloudSat, and (d) merged clouds, that is, clouds detected by CALIPSO or CloudSat. (e) CERES–CALIPSO–CloudSat–MODIS (CCCM)-merged 0.64-µm cloud extinction profiles derived at the CERES footprint scale (∼20 km). (f) A downlooking view of 0.64-µm COT constructed with the method described in Barker et al. (2011). The C-130 flight track is shown in (a) and (b) by the black lines for the entire pattern and by the dashed red line for the period collocated to CERES observations with a time difference <30 min and a distance <20 km. MODIS-derived cloud-top and effective heights are shown in (d) by red and blue dots, respectively. The P and Q markers in (a) correspond to the locations shown of 15 Sep in Fig. 7c.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
Each of the three sets of CERES level 3 evaluation experiments were performed over different surface conditions: over open ocean (15 and 17 September), over the marginal ice zone (MIZ; 7 and 9 September), and over an area of high sea ice concentration (11 and 13 September). All three regions were well sampled, with at least four satellite overpasses (from a combination of Terra, Aqua, and SNPP) during each 2.5–3-h aircraft flight. Figures 7d–f show the distribution of instantaneous CERES-derived SW and LW irradiances at TOA from within each of the orange grid boxes that bound the flight pattern. The distributions of LW and SW irradiances are noticeably different for each of the days. The differences can largely be understood by the cloud and surface conditions present in each of the grid boxes. On 7 September, the surface consisted of marginal ice and open ocean with a very low and quite optically thin overcast cloud layer. This results in a SW irradiance distribution that is skewed toward lower values with the long tail toward higher values due to the marginal sea ice and some cloud optical depth variability. Because the cloud tops were so low, there is little variation in the emission height, resulting in a narrow LW irradiance distribution. On 11 September, the surface consisted of high sea ice concentration with a combination of clear sky and low thin clouds. This creates a bright scene and correspondingly higher SW fluxes. The low cloud tops and cold sea ice results in a narrow LW irradiance distribution. While the surface on 15 September was open ocean, the cloud conditions were overcast, high, and very optically thick (see Fig. 9). This results in the comparatively high SW and low LW fluxes shown in Fig. 8f. These distributions will be compared with the broadband radiometer (BBR) irradiance measurements obtained from the C-130 (with suitable atmospheric correction). BBR irradiances taken near the surface will be compared with computed irradiances from the SYN product. The spectral surface albedo derived from the Solar Spectral Flux Radiometer (SSFR) will be used to evaluate the surface albedo used in the computations.
BBR.
BBRs were mounted on the top and bottom of the aircraft to measure the down- and upwelling global solar (SW) irradiance (0.2–3.6 µm); the downwelling global, direct, and diffuse SW irradiance (0.4–42 µm); and the down- and upwelling infrared (LW) irradiance (4.5–42 µm; see Table 1). Kipp & Zonen pyranometers (Kipp & Zonen 2004) and pyrgeometers (Kipp & Zonen 2001), modified to make them better suited for use on an aircraft, measured the SW and LW irradiances. Modifications included new hermetically sealed back housings with the connector on the bottom that prevented condensation and freezing inside the domes and simplified the mounting of the sensors to the aircraft. The front-end optics and electronics of the original instruments were retained but an amplifier was added right below the sensors and the instruments were operated in current loop mode, a well-established technique to minimize electronic noise.
A Delta-T Devices sunshine pyranometer (SPN-1) was mounted on top of the aircraft to measure the downwelling global, direct, and diffuse SW irradiance. To accomplish this, the SPN-1 has a custom-designed hemispheric “shadowmask” that lies just under the protective glass dome that covers the instrument’s seven thermopile sensors, each topped with a cosine-corrected diffuser and each with a spectral bandpass of 0.4–2.7 µm. The shadowmask is designed to ensure that at least one sensor is always exposed to the direct solar radiation, and at least one sensor is always shaded from the direct beam, independent of the orientation of the instrument to the sun. The global, direct, and diffuse SW irradiances are then derived from these maximum and minimum readings (Delta-T Devices 2007). Although there is some uncertainty regarding the absolute accuracy of the SPN-1 (Badosa et al. 2014), these data are particularly useful to obtain the direct–diffuse ratio needed to correct the downwelling SW irradiances for the attitude of the aircraft (Long et al. 2010; Bucholtz et al. 2008).
The SW radiometers were calibrated using the standard alternating sun–shade method (ASTM 2005), where the given sensor is compared to the true direct solar irradiance measured by an Eppley automatic Hickey–Frieden (AHF) absolute cavity radiometer. The sensitivities for the SW radiometers from pre- and postmission calibrations agreed to within 1%. The LW radiometers were calibrated by comparison of the measured signals to the irradiance of a blackbody immersed in a variable temperature alcohol bath. The calibration coefficients for the LW radiometers from pre- and postmission calibrations agreed to within 2%. Thus, the stability of the SW and LW radiometers during ARISE was excellent. For the SPN-1 the calibration from the manufacturer was used (8% estimated accuracy). This is sufficient here, since the SPN-1 measurements will be mainly used to correct the downwelling BBR SW irradiances for the attitude of the aircraft, which requires only the relative values of the global, direct, and diffuse SW irradiance.
Figures 10a and 10b show the CERES lawn-mower pattern flown on 7 September overlaid on the NOAA-19 red–green–blue (RGB) and IR satellite images taken during the flight at 2150 UTC. A uniform, optically thin low-level cloud deck blanketed the area. The pinker area, apparent in the RGB image of the southeastern half of the pattern, indicates heavy concentrations of sea ice, while the darker areas in the northwestern half of the box indicate mostly open ocean beneath the clouds. The infrared image (Fig. 10b) indicates that the area was mostly clear of high clouds, although some thin scattered cirrus are seen in the northwestern portion of the box. These conditions were confirmed by the onboard flight scientist’s notes and the forward video on the aircraft. Figure 10c is an image grab from the forward video taken at approximately the midpoint of the first leg of the pattern, showing the mostly clear skies aloft and a uniform low-level cloud deck. Figure 10d shows the order in which the lawn-mower pattern was flown. This flight is a good case for comparisons between the CERES and BBR SW and LW irradiances because, while there was some variation in the cloud and surface properties within the box, they remained nearly constant while the aircraft sampled the area. In fact, a particular advantage in conducting this type of experiment in the Arctic in late summer/early fall is that the sun, though low in the sky, remains at a nearly constant elevation angle and thus the incoming solar irradiance at the TOA is nearly constant for a long time during the day. Figure 10e shows that the solar zenith angle θo remained nearly constant (average θo = 69.75° ± 0.62°) during the entire pattern. This simplifies the interpretation of the aircraft irradiances, which take about 2 h to survey over the region, when compared to the nearly instantaneous CERES satellite measurements.
Example of BBR data collection parameters on 7 Sep 2014. Aircraft flight pattern overlaid on the NOAA-19 satellite (a) RGB and (b) IR images from 2150 UTC, (c) an image from the forward video camera, (d) the order of the lawn mower flight pattern, and (e) SZA.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
The corresponding BBR LW and SW irradiances are shown in Fig. 11. Figure 11a shows the measured down- and upwelling LW irradiances. The data during turns has been removed. Little variation in the down- or upwelling LW irradiances from leg to leg is apparent during the pattern. The mean downwelling LW for is 70.17 ± 5.74 W m−2, while the average upwelling LW is 251.90 ± 4.60 W m−2, confirming the uniformity of the conditions with respect to LW irradiance. Figure 11b shows the measured down- and upwelling SW irradiances. The downwelling SW fluxes require correction for the attitude of the aircraft because changes in the pitch, roll, or heading of the aircraft can cause changes in the zenith angle of the sun with respect to the SW radiometer on top of the aircraft. This causes artificial offsets in the downwelling SW measurements (Bucholtz et al. 2008). This can be seen in Fig. 11b for the uncorrected downwelling SW irradiances shown in black. Dramatic shifts in the data are seen from one leg to the next as the aircraft changes heading. Using the pitch, roll, and heading from the aircraft’s navigational system, the downwelling SW fluxes are corrected back to the true solar zenith angle and are found also to remain fairly constant during the flight, as shown in red in Fig. 11b. In this case, the SW irradiances are normalized to the mean solar zenith angle during the pattern (θo = 69.75°) to make the SW measurements consistent throughout the flight pattern. In future analyses, other solar zenith angle (SZA) normalization strategies will be employed (e.g., to the CERES observation time). Most of the variability in downwelling SW is attributed to the scattered thin cirrus that occasionally occurred overhead. The mean downwelling SW irradiance is 399.35 ± 16.87 W m−2. The upwelling SW irradiances show more variation, with increases or decreases within a given leg. This is attributed to the change in the sea surface conditions beneath the low-cloud deck. For example, the upwelling SW irradiances shown in Fig. 11b are smaller at the northwestern end of each leg because of the darker ocean compared to the brighter surfaces found over the southeastern end, where there was much more sea ice. The average upwelling SW irradiance for the entire pattern was 207.33 ± 32.48 W m−2. The upwelling SW and LW irradiances are consistent with earlier Arctic aircraft campaigns (Curry and Herman 1985; Herman and Curry 1984; Pinto 1998; Curry et al. 2000), while the downwelling LW irradiance is smaller due to the aircraft’s higher altitude during this particular flight pattern. This initial analysis is encouraging and supports the sampling strategy devised and employed during ARISE for evaluating CERES TOA and surface irradiances over the Arctic with aircraft measurements. More detailed analyses and comparisons between BBR and CERES are planned for all of the ARISE gridbox experiments.
BBR (a) LW and (b) SW fluxes from the flight tracks in Fig. 10.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
SSFR.
The SSFR (Pilewskie et al. 2003) measures downwelling (zenith: 
The instrument consists of two light collectors at the top and bottom of the aircraft fuselage, as well as a rack-mounted radiometer unit that is connected to the light collectors through fiber-optic bundles. For ARISE, the zenith light collector was mounted on an active leveling platform to keep the receiving plane of the light collectors aligned with the horizon during attitude changes of the airplane. The radiometer box contains two identical pairs of grating spectrometers covering the spectral range: (a) 350–1000 nm (Zeiss grating spectrometer with silicon linear diode array) and (b) 950–2200 nm (Zeiss grating spectrometer with InGaAs linear diode array). More instrument details can be found in Wendisch et al. (2013, chapter 7). The radiometric and angular responses were determined in the laboratory before and after the field deployment; the drift of the radiometric calibration was tracked with a portable field calibrator over the course of the mission (accuracy of 3%), and the horizontal alignment of the leveling platform was adjusted before each flight (accuracy of 0.2°). Because of the low sun elevation in the Arctic, minor misalignments of the instrument with respect to the horizon increase the absolute uncertainty (Wendisch et al. 2001) and low signal levels lead to elevated noise. In addition, reflections and obstructions from the aircraft itself or other instruments affect the measurements under these conditions. Overall, the absolute uncertainty was increased to about 7% for θo < 75°.
SSFR-derived cloud properties for a case on 19 Sep 2015: A cloud field across the MIZ from open to ice covered (small and large symbols, respectively). Green: cloud albedo (F↑/F↓ above the cloud); red: surface albedo (F↑/F↓ below the cloud); blue: transmittance 
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
This example begs the question whether such three-dimensional cloud effects remain significant when averaging over larger domains. A further interesting question concerns the relative magnitude of cloud and water vapor absorption for different types of clouds (thermodynamic phase and altitude) above different surface types. In our example, the water vapor absorption features (relative to the negative baseline caused by horizontal photon transport) are much more prominent than the weak cloud absorption features. For high clouds, the situation may be reverse. This will be quantified in future work, using spectral partitioning of the absorption by constituents (Kindel et al. 2011).
SSFR data will provide spectral surface albedo as a boundary condition for satellite and airborne remote sensing—a first example is shown in Fig. 12. From the measured albedo, transmittance, and absorptance spectra, cloud properties (optical thickness, thermodynamic phase, effective radius) can be derived that are averaged over the SSFR hemispherical footprint. These can be compared with satellite retrievals. The collection of aircraft and satellite cloud retrievals, in situ measurements, and spectral and broadband irradiances is expected to lead to a deeper understanding of the radiative effects of clouds in the MIZ.
4STAR
The Spectrometer for Sky-Scanning, Sun-Tracking Atmospheric Research (4STAR) instrument combines airborne sun tracking and sky scanning with spectroscopy by incorporating a sun-tracking–sky-scanning–zenith-pointing head with fiber-optic signal transmission to rack-mounted grating spectrometers (Dunagan et al. 2013) that cover the ultraviolet–visible (210–995 nm, spectrometer I) and SW infrared (950–1703 nm, spectrometer II) spectral regions, with a spectra acquisition rate of 1 Hz. During ARISE, 4STAR was operated in its three operation modes: sun tracking, sky scanning, and zenith pointing. The 4STAR tracking head was installed in a modified escape hatch in the zenith port at flight station 220 on the NASA C-130. The data acquisition, motion control, and spectrometers were installed further aft at a flight operator station.
In sun-tracking mode, two motors and a quadrant photodiode detector provide active tracking of the solar disk for measurements of direct solar beam transmittance. Dark counts are measured every 20 min with a shutter mechanism. Atmospheric transmittance is derived by dividing the dark-subtracted photon counts by a TOA reference spectrum, accounting for measurement integration time. The TOA reference spectrum is determined by the refined Langley plot method (Shinozuka et al. 2013). In ARISE, we obtained the 4STAR TOA calibration spectrum (Segal Rosenhaimer et al. 2014) using measurements from a dedicated high-altitude flight on 2 October. Direct sun products include aerosol optical depth (AOD; Shinozuka et al. 2013), total column water vapor (CWV), O3, and NO2 (Segal Rosenhaimer et al. 2014) under clear sky and cirrus optical depth under thin cirrus cases (Segal Rosenhaimer et al. 2013).
In sky-scanning mode, 4STAR measures the diffuse sky radiance at prescribed scattering angles from the sun in the almucantar or principal plane to retrieve aerosol properties (single-scattering albedo, size distribution, and refractive index; see Kassianov et al. 2012). In ARISE, a special modification of this mode was applied under cloudy scenes, with the goal of extracting scattering phase function properties from the various cloud types.
In the zenith mode, the instrument points in the zenith direction and measures diffuse radiances, for the retrieval of cloud phase, optical depth, and effective radii, following the method of LeBlanc et al. (2015). This mode is used under cloudy skies and accounts for 18% of the data collected by 4STAR during ARISE. Figure 13 shows an example illustrating the sensitivity to the zenith radiances to cloud optical properties. Modeled radiances closely match the two example measured spectra, with small differences owing to the possible inclusion of cloud particles of mixed phase. The sky radiance measurements were calibrated before and after the 4STAR ARISE deployment to a National Institute of Standards and Technology (NIST)-traceable integrating sphere at the NASA Ames Research Center, and throughout the deployment with a field-portable 15.24-cm-diameter integrating sphere referenced against the same NIST-traceable source.
4STAR spectra of zenith radiance transmitted through cloud over open water and over sea ice (black) on 19 Sep 2014 compared with radiative transfer model simulations for various single-phase clouds (colors). Radiative transfer calculations used the scattering phase functions for ice particles described by Baum et al. (2011) and flight-level spectral albedo measured by SSFR. The green and red curves show the extremes of considered cloud properties, while blue and purple curves represent modeled radiances matching closely measured radiances. Gray areas indicate wavelength regions where the spectral shape can be used to retrieve cloud properties (optical depth, effective particle radius, thermodynamic phase).
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
Cirrus cloud optical thickness (COT) was calculated based on the method detailed in Segal Rosenhaimer et al. (2013). This retrieval approach is based on the generation of lookup tables (LUTs) of total transmittance for the sun photometer’s field of view (FOV) due to the direct and scattered irradiance over the spectral range measured, for a range of cirrus COT (0–4), and a range of ice cloud effective diameters (10–120 µm) by using explicit cirrus optical property models from Baum et al. (2011). To calculate the total transmittance seen by the instrument, which includes both the direct and forward-scattered components, we use a function suggested by Shiobara and Asano (1994), generated by a three-dimensional (3D) Monte Carlo radiative transfer model. Our measurements are then corrected for the appropriate gas absorption and solar zenith angle at the time of measurement and compared to the modeled values over a range of wavelengths, spanning both visible and infrared spectrometers, and are chosen by the best-fit approach. Cirrus locations were adjusted from aircraft coordinates, since the 4STAR tracks the sun and does not view the clouds in zenith directly above the aircraft. The new location was calculated based on distance derived by the estimated cirrus height, the solar zenith angle, and the sun azimuth. Cirrus top height was approximated from MODIS and was ∼9 km (300 mb for cirrus top height). The latter adjusted cirrus location is about 8 km from the aircraft coordinates.
The 4STAR cirrus retrievals will not only aid the interpretation of the aircraft irradiance measurements but also be useful for validating satellite cloud property retrievals, such as COT (by direct comparison), and cloud-top height (CTH; indirectly). For example, Fig. 14 shows the CTH derived from the MODIS imager on Terra at 2140 UTC 15 September 2014. Two sets of MODIS CTH retrievals are shown. The first, shown in Fig. 14a, is based on a single-layer (SL) cloud assumption for all ice-phase clouds (Minnis et al. 2011), which often underestimates CTH (Chang et al. 2010a, and references therein). The second is based on a multilayer (ML) cloud algorithm (Chang et al. 2010b) and shown in Fig. 14b for the upper layer (Fig. 14b). While the satellite CTH estimates from the SL method are near or below the altitude of the aircraft, the upper-level CTHs determined from the ML algorithm are consistently higher than the aircraft, which is corroborated by numerous 4STAR observations of overhead cirrus. Figure 14c shows the total column COT derived from the MODIS SL method. For areas with clouds beneath the aircraft these retrievals are not comparable to 4STAR, since 4STAR is pointing at the overhead sun. However, the MODIS ML COT retrieval for the upper-level cloud should be more comparable to the 4STAR retrievals of cirrus above the aircraft. This statistical comparison is shown in Fig. 15. The CERES–MODIS upper-layer COTs, derived from the 2140 UTC Terra overpass, were spatially interpolated to match the 4STAR cirrus locations (found between 2100 and 2200 UTC). While the overall mean and median values of COT along the flight track are found to agree quite well as shown in Fig. 15 (0.84 and 0.77 for 4STAR and MODIS, respectively), the 4STAR data suggest more widespread cirrus may have been occurring than were detected with the CERES–MODIS method. Only 164 valid CERES points were found in comparison to 664 from 4STAR. One possible contributing factor to this difference is the relatively large 4STAR instrument FOV (compared to MODIS), which spans about 2°, allowing for coverage of the entire solar disk plus about 0.5° from each side. Thus, as the box plots indicate, 4STAR appears to be more sensitive to the optically thinner cirrus clouds, which are difficult to detect from MODIS. A comparison between only the coincident positive cirrus COT retrievals (not shown) indicates that the MODIS mean value of 0.77 is considerably lower than the mean value of 1.3 found from the corresponding 4STAR points. This is useful information that can be used to improve the skill of the satellite method. Table 4 describes the full range of 4STAR data products that will be available from ARISE.
4STAR retrievals of cirrus COT on 15 Sep 2014 compared with MODIS retrievals from the nearest satellite overpass. (a) All ice clouds’ top heights derived by CERES, overlaid by aircraft altitude (open circles). (b) Top-layer cloud height of ML clouds, derived by CERES, overlaid by aircraft altitude (open circles) for 15 Sep flight. (c) COT for all ML ice clouds with top above 5-km height derived by CERES (solid circles), overlaid by direct sun cirrus retrievals [based on procedure developed in Segal-Rosenheimer et al. (2013)] from the 4STAR instrument on board C-130 (open circles). (d) COT for only upper-layer clouds, as derived by CERES, overlaid by direct sun cirrus retrievals from 4STAR (open circles) for 15 Sep flight. Note the different color bar scales for (c) and (d).
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
Cirrus COT statistics from CERES–MODIS upper-layer cloud retrievals (yellow) and those derived from 4STAR (blue) using data taken on 15 Sep 2014 along the C-130 flight track shown in Fig. 14. Solid black lines indicate the median values, while the top and bottom numerical values indicate the mean values and the number of samples, respectively.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
4STAR data products during ARISE.
LARGE cloud probes.
Cloud droplet microphysical properties were measured in situ by the C-130 using multiple probes operated by the NASA Langley Aerosol Research Group (LARGE). The probes were mounted on the starboard side of the aircraft just forward of the propeller line (Fig. 1).
The multielement water content system (WCM-2000; SEA Inc.) is a three-wire probe based on commonly used and proven technologies that are combined to measure the total and liquid water content (TWC and LWC, respectively) simultaneously. The ice water content (IWC) is inferred from the difference between TWC and LWC. During ARISE, most of the mass measured with this instrument was liquid. Typically the ratio LWC/TWC is on the order of 90%–95%, and this is consistent with an earlier aircraft study of autumnal Arctic clouds sampled approximately a month later in the season (Pinto 1998). Uncertainties of 20% have been found across different Johnson–Williams LWC probes in a wind tunnel testing (Strapp and Schemenauer 1982), which lend support to the premise that these are supercooled liquid rather than ice clouds. In our preliminary inspection of the dataset, there does not seem to be a dependence of LWC/TWC across surface types.
Cloud droplet number and size distribution (2–50-μm diameter) are measured with a cloud droplet probe [CDP; Droplet Measurement Technologies (DMT)]. The CDP measures the forward-scattered light from cloud particles that pass through a laser beam. The intensity of the scattered light is related to the cloud particle size assuming spherical particles and is verified using NIST-traceable glass spheres (Thermo Fisher Scientific, Inc.). Liquid water and water vapor path above the aircraft are measured using a G-band (183 GHz) water vapor radiometer (GVR; ProSensing, Inc.). The GVR measures the brightness temperature of four receiver channels centered on the water vapor absorption line at 183.31 ± 1, ±3, ±7, and ±14 GHz. Two internal references (i.e., hot and warm targets at 333 and 293K, respectively) are used to calibrate the receivers once every 10 s during flight.
Low-level Arctic stratus clouds were sampled in situ during each of the ARISE science flights and were consistently observed within the shallow boundary layers spanning 0–350 m in altitude. An example of this vertical structure is shown in Fig. 16 for the research flight on 15 September. For this flight, the C-130 initially transited northwest toward the sea ice edge at approximately 7000 m before descending to the surface to profile three cloud layers centered at approximately 5500, 4000, and 300 m. The aircraft then ascended and descended through the low-cloud layer, for which vertical profiles are shown in Fig. 16, indicating that the cloud layer extended from 30 to 90 m at cloud base up to 490–550 m at cloud top. Mean droplet number concentrations were observed to be relatively constant throughout the cloud layer at approximately 100 cm−3, while both droplet mean diameters and liquid water content increased with altitude (from 4 to 14–16 µm and from 0.15 to 0.4–0.5 g m−3, respectively). Despite being near the monthly mean sea ice extent for September 2014, it was noted at the time that these aircraft maneuvers were conducted over a mostly sea ice–free surface with only the occasional patch of broken sea ice below.
Vertical profiles of cloud microphysical properties for three cloud penetrations on 15 Sep 2014. (left) The location of each profile is highlighted (red: 75.60°N, 156.04°W, blue: 74.82°N, 155.43°W, gold: 76.33°N, 156.83°W). The complete flight track is shown in black, the National Snow and Ice Data Center (NSIDC) monthly mean sea ice extent is shown as solid white, and the MODIS visible imagery is shown for the non-ice region. (top right) Traces of droplet number density, mean droplet size, and LWC at 1-Hz resolution during each profile. (bottom right) Droplet number size distributions binned by altitude.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
This low-cloud structure contrasts that seen for a cloud sampling pattern carried out on 19 September considerably to the east of that on 15 September, where the aircraft flew vertically stacked legs across the sea ice edge from approximately 136° to 129°W longitude. As shown in the top-left panel of Fig. 17, the aircraft initially ascended from west to east while skirting the ever-increasing top of the cloud layer (black trace), then retraced its transect from east to west in a descending/ascending porpoise maneuver (red, blue, gold), and finally turned back west to east for a low-level horizontal leg through the lowest portion of the cloud. The western portion of the low-cloud layer (gold traces in Fig. 17) spanned 150–1200 m altitude, while the eastern portion (red traces) was shifted higher (600–2100 m). Despite these differences, typical droplet number concentrations, mean droplet diameter, and liquid water content were of similar magnitude across all three profiles.
Sampling of cloud properties across the ice edge centered near 72.3°N, 133.5°W on 19 Sep 2014. (top left) The altitude vs longitude trace shows the aircraft sampling strategy along (bottom left) the parallel tracks. Initially, the aircraft ascended from west to east following just above the cloud top (black trace). Then, three vertical profiles were carried out to map the vertical extent of the clouds (red, blue, gold). The vertical profile of droplet number density, mean diameter and LWC are shown in the lower right. Finally, a series of horizontal legs was performed at 800 ft (∼245 m) along the same track. The cloud properties along one of these legs is shown in the upper right. Green (cyan) denotes the underflight of the gold (blue) profiles at 800 ft.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
In addition to the vertical cloud structure, level flight legs (green and cyan in Fig. 17) show a marked amount of horizontal variability. The cloud droplet number concentration and LWC traces in the top-right panel of Fig. 17 show an alternating pattern of cloud and cloud-free air as both LWC and the cloud droplet number concentration (CDNC) drop quickly to zero for brief periods of time. This cloud structure was clearly visible from the aircraft during this (and other) flight—the ocean surface could be discerned when looking at angles near nadir, while the view at lower angles was entirely opaque. Finally, we note the strong increase in cloud droplet number and corresponding decreases in both LWC and mean droplet diameter as the aircraft passed over the ice sheet edge. This transition may be explained by a shift in the dynamics controlling these clouds or, possibly, by an increase in cloud condensation nuclei over the open waters.
LVIS.
NASA’s Land, Vegetation and Ice Sensor (LVIS) is a wide-swath scanning laser altimeter (lidar) system that digitally records the shapes of the outgoing and reflected laser pulses (Blair et al. 1999). Information extracted from the laser waveforms is combined postflight with precise laser pointing, scanning, and positioning data to precisely and accurately measure surface elevation and 3D surface structure relative to a reference surface, such as the World Geodetic System 1984 (WGS-84) reference ellipsoid (Hofton et al. 2008). Operating at a wavelength of 1064 nm and at a data rate of 1500 Hz, typical data precision and accuracy are at the 10-cm level over ice surfaces (Hofton et al. 2008). The sensor is used to collect data for cryospheric, ecological, biodiversity, and solid-Earth applications, providing a characterization of the three-dimensional nature of overflown surfaces. An atmospheric channel, implemented for the first time for the ARISE mission, provided a record of the returns at 1064 nm along the full laser path from the airplane to the ground. During data processing, these waveforms were combined over 1-s intervals within a common elevation range to provide the vertical distribution of reflected surfaces between the laser and the ground.
During ARISE, the sensor operated in two principal configurations that defined the data swath width. From medium to high flight altitudes, the full laser swath width was used. For example, from a 7-km flight altitude the laser swath was ∼1400 m wide with an 18-m-wide footprint. From lower altitudes, in order to prevent overstressing of system components, an 80-mrad-wide laser swath was used (e.g., from a 0.45-km flight altitude, the laser swath was ∼4.5 m wide with eight ∼1-m-wide footprints). Data products include the geolocated return laser waveform, defining the vertical distribution of the reflecting surfaces within the laser footprint relative to the reference ellipsoid (level 1B), and elevation data products extracted from the level 1B laser waveform using standard waveform interpretation algorithms, in this case the locations of the lowest and highest reflecting surfaces with the laser footprint (level 2).
Data were typically collected throughout each ARISE flight even if the surface was not discernible through clouds in order to enable both radiation and ice target objectives to be met. Mission highlights included a 1,000-km-long transect from open water to sea ice along the 140°W longitude line (Fig. 18); a 600-km-long transect of an orbit track of the European Space Agency (ESA)’s Cryosat-2 with the satellite passing directly overhead at the start of the line; repeated passes over the MIZ throughout the ARISE campaign over the time of the sea ice minimum; data swaths along several Alaskan glaciers, including the Columbia, Portage, Spencer, Trail, and Wolverine glaciers; and characterization of cloud-top heights throughout each flight to interpret the radiation measurements (Fig. 19). The LVIS team is developing a cloud-top height product based on the laser returns. As long as the laser beam is not fully attenuated, there is information on the top height of multiple cloud layers.
Surface elevation data derived from the scanning LVIS superimposed on digital camera imagery taken near 76.4°N, 140°W on 5 Sep 2014 to help characterize sea ice properties and variability.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
LVIS return pulse intensity (a) along the C-130 flight track and (b) on 15 Sep 2014, depicting surface and cloud-top altitudes. An LVIS level 2 cloud altitude product is in development to complement and help interpret the cloudy-sky radiative flux measurements obtained during ARISE.
Citation: Bulletin of the American Meteorological Society 98, 7; 10.1175/BAMS-D-14-00277.1
SUMMARY AND FUTURE WORK.
ARISE was a uniquely successful experiment in three respects. First, the experiment collected advanced radiometric, laser altimeter, and in situ atmospheric data during the critical period of late summer and early autumn sea ice transition in the Beaufort Sea. Second, the aircraft measurements were effectively coordinated with multiple intersecting satellite overpasses, allowing for thorough validation of CERES climate data record products plus a greater understanding of the subgrid-scale variability that influences satellite products at high northern latitudes. Third, the experiment was conceived, planned, and executed in a remarkably short time—6 months from concept to flight missions, whereas many other experiments of this complexity often take several years to realize. This third success also entails a challenge for the ARISE Science Team: our expertise is almost exclusively within the domains of the flight instruments and data interpretation specific to the instruments and satellite remote sensing. We therefore invite and encourage as wide a collaboration as possible with the broader community, particularly researchers interested in 1) applying the resulting well-tested CERES data products to global and regional climate modeling and climate change studies and 2) applying the combination of spectrally resolved and radiometric data and sea ice structure data to process studies involving radiant ice–ocean–atmosphere energy exchange during the sea ice transition. Already we have noticed one potentially important aspect of the clouds sampled throughout ARISE: there is a pronounced tendency toward liquid water in lower- and midtropospheric clouds, with relatively little radiative influence of cloud ice particles as compared with the geometrically extensive mixed-phase clouds observed over the region later in autumn (Verlinde et al. 2007). In this sense the cloud cover during the critical sea ice transition may be more typical of summer (e.g., Tjernström et al. 2012) than autumn. This merits further investigation because ice water content in Arctic mixed-phase clouds exerts a significantly contrasting radiative forcing compared with clouds that are almost entirely liquid water (Lubin and Vogelmann 2011). At the same time, the apparent simplicity of a cloud possibly dominated by a single thermodynamic phase may be offset by the 3D radiative transfer effects noted above (Fig. 12), and the high-time-resolution spectral radiometric data from ARISE can address these complexities.
The ARISE data, which are available at the NASA Langley Atmospheric Science Data Center and in the NASA OIB archive, contain a wealth of information on the Arctic sea ice transition from in situ process to satellite spatial scales. In addition to data analysis from the campaign itself, ARISE can help motivate future work. The average September Arctic sea ice extent exhibits large interannual variability of approximately 1,000,000 km2, in addition to the pronounced downward trend over the past three decades (Stroeve et al. 2012). Additional missions during this transition season with similar instrumentation could provide insight into the precise radiative and thermodynamic precursors for onset of seasonal ice recovery. Stroeve et al. (2014) show that the timing of the melt onset impacts the amount of insolation absorbed during summer, which in turn influences the timing of the autumn ice recovery. Similar attention, perhaps an additional campaign, should focus on the springtime melt onset in the Beaufort Sea. Finally, for both the satellite and in situ objectives presented here, a follow-on aircraft mission would benefit from additional active sensors, such as polarized cloud lidar and cloud radar; a more complete cloud microprobe suite, including aerosol composition and microphysics; and dropsondes, to provide measurements of atmospheric thermodynamic structure specifically over ice of varying concentrations versus open water during a given mission. ARISE has demonstrated what is possible from long-range research aircraft; over the next decade, enhancements to instrumentation combined with a focus on timing of sea ice melt onset and autumn recovery can provide a foundation for thorough understanding of mechanisms for Arctic sea ice trends.
ACKNOWLEDGMENTS
ARISE was sponsored and supported by the Earth Science Division of NASA’s Science Mission Directorate. We thank the program managers at NASA headquarters: Jack Kaye, Hal Maring, Bruce Tagg, and Tom Wagner. Their support and inspiration were critical in the planning and successful execution of ARISE given the urgency for the mission and the short time frame to accomplish it. We are grateful to the personnel at the NASA Wallops Flight Facility, who provided support for the C-130. We thank the C-130 flight crew and integration engineers for their support and significant accomplishes in readying and maintaining the aircraft. We particularly thank the pilots of the C-130—John Long and Jeff Chandler—for their expertise and genuine interest in helping us to accomplish our science objectives. We also thank the NASA Ames Earth Science Project Office for a number of contributions, including logistics support; the National Suborbital Education and Research Center (NSERC) for its support of the aircraft data system; Aaron Duley and his colleagues at the NASA Ames Research Center for configuring and maintaining the NASA Mission Tools Suite; and Nathan Eckstein and his colleagues at the Alaska Aviation Weather Unit for providing valuable meteorological support. Finally, we are grateful to the staff members at Eielson Air Force Base for all of their support in hosting the C-130 and the ARISE experiment team.
REFERENCES
ASTM, 2005: Standard test method for calibration of a pyranometer using a pyrheliometer. ASTM G167–05, ASTM International, 21 pp, doi:10.1520/G0167-05.
Badosa, J., J. Wood, P. Blanc, C. N. Long, L. Vuilleumier, D. Demengel, and M. Haeffelin, 2014: Solar irradiances measured using SPN1 radiometers: Uncertainties and clues for development. Atmos. Meas. Tech., 7, 4267–4283, doi:10.5194/amt-7-4267-2014.
Barker, H. W., M. P. Jerg, T. Wehr, S. Kato, D. P. Donovan, R. J. Hogan, 2011: A 3D cloud-construction algorithm for the EarthCARE satellite mission. Q. J. R. Meteor. Soc., 137, 1042–1058, doi:10.1002/qj.824.
Baum, B. A., P. Yang, A. J. Heymsfield, C. G. Schmitt, Y. Xie, A. Bansemer, Y. X. Hu, and Z. Zhang, 2011: Improvements in shortwave bulk scattering and absorption models for the remote sensing of ice clouds. J. Appl. Meteor. Climatol., 50, 1037–1056, doi:10.1175/2010JAMC2608.1.
Blair, J. B., D. L. Rabine, and M. A. Hofton, 1999: The Laser Vegetation Imaging Sensor (LVIS): A medium-altitude, digitization-only, airborne laser altimeter for mapping vegetation and topography. ISPRS J. Photogramm. Remote Sens., 54, 115–122, doi:10.1016/S0924-2716(99)00002-7.
Bucholtz, A., R. T. Bluth, B. Kelly, S. Taylor, K. Batson, A. W. Sarto, T. P. Tooman, and R. F. McCoy Jr., 2008: The Stabilized Radiometer Platform (STRAP)—An actively stabilized horizontally level platform for improved aircraft irradiance measurements. J. Atmos. Oceanic Technol., 25, 2161–2175, doi:10.1175/2008JTECHA1085.1.
Cassano, E. N., J. J. Cassano, M. E. Higgins, and M. C. Serreze, 2014: Atmospheric impacts of an Arctic sea ice minimum as seen in the Community Atmosphere Model. Int. J. Climatol., 34, 766–779, doi:10.1002/joc.3723.
Chang, F.-L., P. Minnis, B. Lin, M. Khaiyer, R. Palikonda, and D. Spangenberg, 2010a: A modified method for inferring upper troposphere cloud top height using the GOES 12 imager 10.7 and 13.3 μm data. J. Geophys. Res., 115, D06208, doi:10.1029/2009JD012304.
Chang, F.-L., P. Minnis, J. K. Ayers, M. J. McGill, R. Palikonda, D. A. Spangenberg, W. L. Smith Jr., and C. R. Yost, 2010b: Evaluation of satellite-based upper troposphere cloud top height retrievals in multilayer cloud conditions during TC4. J. Geophys. Res., 115, D00J065, doi:10.1029/2009JD013305.
Coddington, O. M., and Coauthors, 2008: Aircraft measurements of spectral surface albedo and its consistency with ground-based and space-borne observations. J. Geophys. Res., 113, D17209, doi:10.1029/2008JD010089.
Coddington, O. M., and Coauthors, 2010: Examining the impact of overlying aerosols on the retrieval of cloud optical properties from passive remote sensing. J. Geophys. Res., 115, D10211, doi:10.1029/2009JD012829.
Coddington, O. M., P. Pilewskie, K. S. Schmidt, P. J. McBride, and T. Vukicevic, 2013: Characterizing a new surface-based shortwave cloud retrieval technique, based on transmitted radiance for soil and vegetated surface types. Atmosphere, 4, 48–71, doi:10.3390/atmos4010048.
Curry, J. A., and G. F. Herman, 1985: Infrared radiative properties of summertime Arctic stratus clouds. J. Climate Appl. Meteor., 24, 525–538, doi:10.1175/1520-0450(1985)024<0525:IRPOSA>2.0.CO;2.
Curry, J. A., and Coauthors, 2000: FIRE Arctic Clouds Experiment. Bull. Amer. Meteor. Soc., 81, 5–29, doi:10.1175/1520-0477(2000)081<0005:FACE>2.3.CO;2.
Delta-T Devices, 2007: User manual for the Sunshine pyranometer, type SPN-1. 43 pp. [Available online at ftp://ftp.dynamax.com/manuals/SPN1_Manual.pdf.]
Doelling, D. R., and Coauthors, 2013: Geostationary enhanced temporal interpolation for CERES flux products. J. Atmos. Oceanic Technol., 30, 1072–1090, doi:10.1175/JTECH-D-12-00136.1.
Dunagan, S. E., and Coauthors, 2013: Spectrometer for Sky-Scanning Sun-Tracking Atmospheric Research (4STAR): Instrument technology. Remote Sens., 5, 3872–3895, doi:10.3390/rs5083872.
English, J. M., J. E. Kay, A. Gettelman, X. Liu, Y. Wang, Y. Zhang, and H. Chepfer, 2014: Contribution of clouds, surface albedos and mixed-phase ice nucleation schemes to Arctic radiation biases in CAM5. J. Climate, 27, 5174–5197, doi:10.1175/JCLI-D-13-00608.1.
Fridlind, A. M., A. S. Ackerman, G. McFarquhar, G. Zhang, M. R. Poellot, P. J. DeMott, A. J. Prenni, and A. J. Heymsfield, 2007: Ice properties of single-layer stratocumulus during the Mixed-Phase Arctic Cloud Experiment: 2. Model results. J. Geophys. Res., 112, D24202, doi:10.1029/2007JD008646.
Fridlind, A. M., B. van Diedenhoven, A. S. Ackerman, A. Avramov, A. Mrowiec, H. Morrison, P. Zuidema, and M. D. Shupe, 2012: A FIRE-ACE/SHEBA case study of mixed-phase Arctic boundary layer clouds: Entrainment rate limitations on rapid primary ice nucleation processes. J. Atmos. Sci., 69, 365–389, doi:10.1175/JAS-D-11-052.1.
Hartmann, D. L., and P. Ceppi, 2014: Trends in the CERES dataset, 2000–13: The effect of sea ice and jet shifts and comparison to climate models. J. Climate, 27, 2444–2456, doi:10.1175/JCLI-D-13-00411.1.
Herman, G. F., and J. A. Curry, 1984: Observational and theoretical studies of solar radiation in Arctic stratus clouds. J. Climate Appl. Meteor., 23, 5–24, doi:10.1175/1520-0450(1984)023<0005:OATSOS>2.0.CO;2.
Hines, K. M., D. H. Bromwich, L. Bai, C. M. Bitz, J. G. Powers, and K. W. Manning, 2015: Sea ice enhancements to Polar WRF. Mon. Wea. Rev., 143, 2363–2385, doi:10.1175/MWR-D-14-00344.1.
Hofton, M. A., J. B. Blair, S. B. Luthcke, and D. L. Rabine, 2008: Assessing the performance of 20–25 m footprint waveform lidar data collected in ICESat data corridors in Greenland. Geophys. Res. Lett., 35, L24501, doi:10.1029/2008GL035774.
Intrieri, J. M., C. W. Fairall, M. D. Shupe, P. O. G. Persson, E. L Andreas, P. S. Guest, and R. E. Moritz, 2002: An annual cycle of Arctic surface cloud forcing at SHEBA. J. Geophys. Res., 107, 8039, doi:10.1029/2000JC000439.
Itterly, K., and P. C. Taylor, 2014: Evaluation of the tropical TOA flux diurnal cycle in MERRA and ERA-Interim retrospective analyses. J. Climate, 27, 4781–4796, doi:10.1175/JCLI-D-13-00737.1.
Kassianov, E., C. Flynn, J. Redemann, B. Schmid, P. B. Russell, and A. Sinyuk, 2012: Initial assessment of the Spectrometer for Sky-Scanning, Sun-Tracking Atmospheric Research (4STAR)-based aerosol retrieval: Sensitivity study. Atmosphere, 3, 495–521, doi:10.3390/atmos3040495.
Kato, S., and Coauthors, 2011: Improvements of top-of-atmosphere and surface irradiance computations with CALIPSO-, CloudSat-, and MODIS-derived cloud and aerosol properties. J. Geophys. Res., 116, D19209, doi:10.1029/2011JD016050.
Kato, S., N. G. Loeb, F. G. Rose, D. R. Doelling, D. A. Rutan, T. E. Caldwell, L. Yu, and R. A. Weller, 2013: Surface irradiances consistent with CERES-derived top-of-atmosphere shortwave and longwave irradiances. J. Climate, 26, 2719–2740, doi:10.1175/JCLI-D-12-00436.1.
Kindel, B. C., P. Pilewskie, K. S. Schmidt, O. Coddington, and M. D. King, 2011: Solar spectral absorption by marine stratus clouds: Measurements and modeling. J. Geophys. Res., 116, D10203, doi:10.1029/2010JD015071.
Kipp & Zonen, 2001: CG4 pyrgeometer: Instruction manual. 64 pp. [Available online at www.kippzonen.com/Download/33/CG-4-Manual.]
Kipp & Zonen, 2004: CM22 precision pyranometer: Instruction manual. 65 pp. [Available online at www.kippzonen.com/Download/55/CM-22-Pyranometer-Manual.]
LeBlanc, S. E., P. Pilewskie, K. S. Schmidt, and O. Coddington, 2015: A spectral method for discriminating thermodynamic phase and retrieving cloud optical thickness and effective radius using transmitted solar radiance spectra. Atmos. Meas. Tech., 8, 1361–1383, doi:10.5194/amt-8-1361-2015.
Loeb, N. G., K. J. Priestley, D. P. Kratz, E. B. Geier, R. N. Green, B. A. Wielicki, P. O’Rawe Hinton, and S. K. Nolan, 2001: Determination of unfiltered radiances from the Clouds and the Earth’s Radiant Energy System instrument. J. Appl. Meteor., 40, 822–835, doi:10.1175/1520-0450(2001)040<0822:DOURFT>2.0.CO;2.
Loeb, N. G., S. Kato, K. Loukachine, and N. Manalo-Smith, 2005: Angular distribution models for top-of-atmosphere radiative flux estimation from the Clouds and the Earth’s Radiant Energy System instrument on the Terra satellite. Part I: Methodology. J. Atmos. Oceanic Technol., 22, 338, doi:10.1175/JTECH1712.1.
Loeb, N. G., B. A. Wielicki, D. R. Doelling, G. L. Smith, D. F. Keyes, S. Kato, N. Manalo-Smith, and T. Wong, 2009: Toward optimal closure of the Earth’s top-of-atmosphere radiation budget. J. Climate, 22, 748–766, doi:10.1175/2008JCLI2637.1.
Loeb, N. G., S. Kato, W. Su, T. Wong, F. G. Rose, D. R. Doelling, J. R. Norris, and X. Huang, 2012: Advances in understanding top-of-atmosphere radiation variability from satellite observations. Surv. Geophys., 33, 359–385, doi:10.1007/s10712-012-9175-1.
Long, C. N., A. Bucholtz, H. Jonsson, B. Schmid, A. Vogelmann, and J. Wood, 2010: A method of correcting for tilt from horizontal in downwelling shortwave irradiance measurements on moving platforms. Open Atmos. Sci. J., 4, 78–87, doi:10.2174/1874282301004010078.
Lubin, D., and A. M. Vogelmann, 2011: The influence of mixed-phase clouds on surface shortwave irradiance during the Arctic spring. J. Geophys. Res., 116, D00T05, doi:10.1029/2011JD015761.
McBride, P. J., K. S. Schmidt, P. Pilewskie, A. Walther, A. K. Heidinger, D. E. Wolfe, C. W. Fairall, and S. Lance, 2012: CalNex cloud properties retrieved from a ship-based spectrometer and comparisons with satellite and aircraft retrieved cloud properties. J. Geophys. Res., 117, D00V23, doi:10.1029/2012JD017624.
Minnis, P., and Coauthors, 2011: CERES Edition-2 cloud property retrievals using TRMM VIRS and Terra and Aqua MODIS data—Part I: Algorithms. IEEE Trans. Geosci. Remote Sens., 49, 4374–4400, doi:10.1109/TGRS.2011.2144601.
Molod, A. M., L. L. Takacs, M. Suarez, and J. Bacmeister, 2015: Development of the GEOS-5 atmospheric general circulation model: Evolution from MERRA to MERRA2. Geosci. Model Dev., 8, 1339–1356, doi:10.5194/gmd-8-1339-2015.
Perovich, D. K., and C. Polashenski, 2012: Albedo evolution of seasonal Arctic sea ice. Geophys. Res. Lett., 39, L08501, doi:10.1029/2012GL051432.
Perovich, D. K., J. A. Richter-Menge, K. F. Jones, and B. Light, 2008: Sunlight, water, and ice: Extreme Arctic sea ice melt during the summer of 2007. Geophys. Res. Lett., 35, L11501, doi:10.1029/2008GL034007.
Pilewskie, P. J., and Coauthors, 2003: Solar spectral radiative forcing during the Southern African Regional Science Initiative. J. Geophys. Res., 108, 8486, doi:10.1029/2002JD002411.
Pincus, R., C. P. Batstone, R. J. P. Hofmann, K. E. Taylor, and P. J. Glecker, 2008: Evaluating the present-day simulation of clouds, precipitation, and radiation in climate models. J. Geophys. Res., 113, D14209, doi:10.1029/2007JD009334.
Pinto, J. O., 1998: Autumnal mixed-phase cloudy boundary layers in the Arctic. J. Atmos. Sci., 55, 2016–2038, doi:10.1175/1520-0469(1998)055<2016:AMPCBL>2.0.CO;2.
Rienecker, M. M., and Coauthors, 2008: The GOES-5 data assimilation system—Documentation of versions 5.0.1, 5.1.0, and 5.2.0. M. J. Suarez, Ed., Technical Report Series on Global Modeling and Data Assimilation, Vol. 27, NASA Tech. Memo. NASA/TM-2008–104606, 101 pp. [Available online at https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20120011955.pdf.]
Rutan, D. A., S. Kato, D. R. Doelling, F. G. Rose, L. T. Nguyen, T. E. Caldwell, and N. G. Loeb, 2015: CERES synoptic product: Methodology and validation of surface radiant flux. J. Atmos. Oceanic Technol., 32, 1121–1143, doi:10.1175/JTECH-D-14-00165.1.
Schmidt, K. S., and P. Pilewskie, 2012: Airborne measurements of spectral shortwave radiation in cloud and aerosol remote sensing and energy budget studies. Light Scattering Reviews 6: Light Scattering and Remote Sensing of Atmosphere and Surface, A. A. Kokhanovsky, Ed., Springer, 239–288, doi:10.1007/978-3-642-15531-4_6.
Schmidt, K. S., and Coauthors, 2010: Apparent absorption of solar spectral irradiance in heterogeneous ice clouds. J. Geophys. Res., 115, D00J22, doi:10.1029/2009JD013124.
Segal Rosenhaimer, M., P. B. Russell, J. M. Livingston, S. Ramachandran, J. Redemann, and B. A. Baum, 2013: Retrieval of cirrus properties by Sun photometry: A new perspective on an old issue. J. Geophys. Res. Atmos., 118, 4503–4520, doi:10.1002/jgrd.50185.
Segal Rosenhaimer, M., and Coauthors, 2014: Tracking elevated pollution layers with a newly developed hyperspectral Sun/Sky spectrometer (4STAR): Results from the TCAP 2012 and 2013 campaigns. J. Geophys. Res. Atmos., 119, 2611–2628, doi:10.1002/2013JD020884.
Shinozuka, Y., and Coauthors, 2013: Hyperspectral aerosol optical depths from TCAP flights. J. Geophys. Res., 118, 12 180–12 194, doi:10.1002/2013JD020596.
Shiobara, M., and S. Asano, 1994: Estimation of cirrus optical thickness from sun photometer measurements. J. Appl. Meteor., 33, 672–681, doi:10.1175/1520-0450(1994)033<0672:EOCOTF>2.0.CO;2.
Song, S., and Coauthors, 2016: The spectral signature of cloud spatial structure in shortwave irradiance. Atmos. Chem. Phys., 16, 13 791–13 806, doi:10.5194/acp-16-13791-2016.
Strapp, J. W., and R. S. Schemenauer, 1982: Calibrations of Johnson-Williams liquid water content meters in a high-speed icing tunnel. J. Appl. Meteor., 21, 98–108, doi:10.1175/1520-0450(1982)021<0098:COJWLW>2.0.CO;2.
Stroeve, J. C., M. C. Serreze, M. M. Holland, J. E. Kay, J. Maslanik, and A. P. Barrett, 2012: The Arctic’s rapidly shrinking sea ice cover: A research synthesis. Climatic Change, 11, 1005–1027, doi:10.1007/s10584-011-0101-1.
Stroeve, J. C., T. Markus, L. Boisvert, J. Miller, and A. Barrett, 2014: Changes in Arctic melt season and implications for sea ice loss. Geophys. Res. Lett., 41, 1216–1225, doi:10.1002/2013GL058951.
Su, W., J. Corbett, Z. Eitzen, and L. Liang, 2015a: Next-generation angular distribution models for top-of-atmosphere radiative flux calculation from CERES instruments: Methodology. Atmos. Meas. Tech., 8, 611–632, doi:10.5194/amt-8-611-2015.
Su, W., J. Corbett, Z. Eitzen, and L. Liang, 2015b: Next-generation angular distribution models for top-of-atmosphere radiative flux calculation from the CERES instruments: Validation. Atmos. Meas. Tech., 8, 3297–3313