The role of shortwave radiation in the 2007 Arctic sea ice anomaly

Transcription

1 GEOPHYSICAL RESEARCH LETTERS, VOL. 39,, doi: /2012gl052415, 2012 The role of shortwave radiation in the 2007 Arctic sea ice anomaly Eric A. Nussbaumer 1 and Rachel T. Pinker 1 Received 18 May 2012; revised 3 July 2012; accepted 9 July 2012; published 14 August [1] Recent satellite and re-analysis model results on downwelling surface shortwave (DSSW) radiation allow the investigation of its role in the Arctic sea ice anomalies. Using satellite based information we revisit claims that reduced cloudiness and enhanced DSSW are associated with the significant loss of sea ice during We account for the fact that the Arctic Ocean is not homogenous in terms of the characteristics of the sea ice anomalies. We separate the Arctic Ocean region according to the spatial distribution of the sea ice anomalies and investigate the impact of DSSW on the ice conditions accordingly. The region which exhibits the strongest signal during the unprecedented 2007 reduction in sea ice is identified as 120 E to 210 E. Our analysis shows that the lowest cloud amount and the highest accumulated amount of DSSW prior to the ice melt season (June) occur in For 2007, areas showing the largest accumulation of DSSW do not correspond with negative sea ice concentration anomalies. Citation: Nussbaumer, E. A., and R. T. Pinker (2012), The role of shortwave radiation in the 2007 Arctic sea ice anomaly, Geophys. Res. Lett., 39,, doi: /2012gl Introduction [2] Changes in the extent of the Arctic sea ice have implications in areas such as climate, ecosystem development, and natural resource availability. For instance, the loss of sea ice opens up the important North-East and North- West shipping passages during the summertime and allows access to the vast reserve of natural resources located beneath the Arctic sea floor. The United States Geological Survey (USGS) estimates that 22% of the world s undiscovered and recoverable oil and natural gas/liquid reserves may be located within the seafloor of the Arctic Region (U.S. Geological Survey, 90 billion barrels of oil and 1,670 trillion cubic feet of natural gas assessed in the Arctic, press release, 2008, ID=1980&from=rss_home). The most immediate ecological changes are occurring for species that depend on sea ice for foraging, reproduction, and predator avoidance [Post et al., 2009]. [3] Since 1979 satellite observations provide a tool to observe changes in the Arctic sea ice extent. One can look at the long term trend (1979 to 2012) of decreasing sea ice extent which follows a slope of 2.6 (+/ 0.6) % per decade as measured by a least squares regression of the ice extent anomalies for each month (The National Snow and Ice Data 1 Department of Atmospheric and Oceanic Science, University of Maryland, College Park, Maryland, USA. Corresponding author: R. T. Pinker, Department of Atmospheric and Oceanic Science, University of Maryland, College Park, MD 20742, USA. (pinker@atmos.umd.edu) American Geophysical Union. All Rights Reserved /12/2012GL Center, NSIDC, or at the seasonal oscillation of melt and freeze based on the monthly sea ice concentration anomaly (a given month s concentration is subtracted from a climatological mean ( )) for that month). [4] In 2007 satellite observations revealed the largest seasonal change in sea ice extent since measurement began in The annual minimum sea ice extent in 2007 was 24% below the previous record set in 2005 and 37% lower than the average [Comiso, 2008]. Several factors are believed to play a role in the seasonal Arctic sea ice anomaly including the changes in the thickness of sea ice [Nghiem et al., 2007], the ice-albedo feedback [Perovich et al., 2007a, 2007b, 2008], Arctic ocean heat transport [Shimada et al., 2006], atmospheric heat transport [Serreze et al., 2007], transport of sea ice through wind stress [Kauker et al., 2009; Lindsay et al., 2009; Rigor and Wallace, 2004; Ogi and Wallace, 2007], downwelling shortwave and longwave radiation [Graversen et al., 2011; Kay et al., 2008; Stroeve et al., 2008; Schweiger et al., 2008]. The relative contribution of each of these factors is unknown and most likely all play a role in modulating the sea ice extent. The focus of this study is to revisit previous claims on the role of clouds and DSSW in the Arctic Sea ice concentration anomalies. We show that over the region of the largest negative sea ice concentration anomaly in 2007, the cloud fraction was not at a minimum and DSSW was not at a maximum as claimed in the study of Kay et al. [2008] and that the lowest cloud fraction and largest DSSW for 2007 occur over a region which shows positive or no sea ice concentration anomaly. 2. Methodology 2.1. Downwelling Surface Shortwave Radiation [5] An inference scheme (DSSW/UMD) for deriving DSSW fluxes that utilizes MODIS Level-3 Atmosphere Daily Global Product information served as a basis for this study [Wang and Pinker, 2009; Pinker et al., 2009]. MODIS information is available on atmospheric constituents, both water and ice cloud properties and surface albedo. The spectral shortwave radiation is retrieved for a multi-layered atmosphere which accounts for surface elevation and for the representation of the vertical distribution of atmospheric variables. [6] Subsequently, the DSSW/UMD model has been updated (v2) to incorporate improved information on surface properties at high latitudes [Niu et al., 2010] such as higher resolution snow cover at 0.05 degree at a daily time scale (MOD10C1 from Terra, MYD10C1 from Aqua) and at monthly time scale (MOD10CM from Terra, MYD10CM from Aqua) available from the National Snow and Ice Data Center (NSIDC) ( index.html). The MODIS surface reflectance products provide a five-year ( ) statistics of spectral reflectance when 1of6

2 Table 1. A summary of the Correlation Statistics Between DSSW Estimated From the DSSW/UMD v2 Model and Two Ground Stationsa Time Scale Daily Monthly Station Correlation NSA NYA NSA NYA Bias [W m 2] RMSE [W m 2] a NSA-North Slope of Alaska ( N, E) (ARM) NYA-NyÅlesund ( N, E) (Baseline Surface Radiation Network). the surface is covered by snow (the underlying surface types are aggregated according to the International GeosphereBiosphere Program (IGBP) classification) [Moody et al., 2007]. For instance, the white-sky reflectance at wavelength of 0.55 mm for snow-covered grassland, desert and permanent snow are 0.72, 0.87 and 0.94 [Moody et al., 2007]. The original scheme [Wang and Pinker, 2009] was implemented with monthly mean sea ice extent at 1 grid cells based on the Special Sensor Microwave/Imager(NOAA/NESDIS). In the updated version used here, 25 km sea ice concentrations as Figure 1. Regions used to determine the relationship between accumulated irradiance and sea ice concentration anomaly in the Arctic Ocean. Region 1: 120 E to 150 E and 70 N to 90 N; Region2: 150 E to 180 E and 70 N to 90 N; Region 3: 180 E to 210 E and 70 N to 90 N; Region 4: 210 E to 240 E and 70 N to 90 N. Arctic Ocean map from org/wiki/arctic_ocean. 2 of 6

3 Figure 2. Arctic sea ice concentration anomalies for the corresponding regions show in Figure 1. The solid black line shows the sea ice concentration anomaly for each region. The dashed black line is the sea ice concentration anomaly for the entire region: 0 E 360 E and 70 Nto90 N. derived from the Nimbus-7 Scanning Multichannel Microwave Radiometer (SMMR) and the Defense Meteorological Satellite Program (DMSP) Special Sensor Microwave/Imager (SSM/I) radiances are used (based on a NASA algorithm [Cavalieri et al., 1996, 2008] as available at daily and monthly time scales) ( The fixed values of spectral reflectance of sea ice in the original scheme have been updated for four distinct phases in the margin areas of the Arctic (winter stationary, spring melt season, summer stationary and autumn freeze up) according to Zhang et al. [2003] and Belchansky et al. [2004]. Since the daily snow cover has many missing values at high latitudes, they have been replaced with monthly mean values. On a daily and monthly timescale, from 2003 to 2007, estimates from DSSW/UMD agree well with the Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) Program ground station data [Dong et al., 2010] as summarized in Table 1. The correlation coefficients have a p-value of 0.000, indicating significance Sea Ice Data [7] Arctic sea ice concentration anomaly is calculated by taking the difference between a monthly climatology of sea ice concentration and the actual monthly concentration. The monthly climatology is a mean ice concentration for each month derived from data from 1979 to The data are derived from the NASA Sea Ice Concentrations from Nimbus-7, the same source as used in the DSSW algorithm and described in section 2.1. The monthly Arctic Sea Ice concentration anomaly is generated using the Advanced Microwave Scanning Radiometer Earth Observing System (AMSR-E) Bootstrap Algorithm with daily varying tie-points. Both data-sets are available from the National Snow and Ice Data Center (NSIDC) Equal Area Comparison [8] When taking averages in the Arctic it is important that the area of each grid cell is properly weighted. In the Polar Regions (60 90 ) the area of a 1 by 1 grid cell can range from 6,123 km 2 (centered at 60.5 N latitude) to 109 km 2 (centered at 89.5 N latitude). One can use a weighted mean, weighted by the area of each grid cell or develop the mean from an equal area grid. The analysis presented here conforms to the latter. 3. Results 3.1. Cloud Fraction and Arctic Sea Ice Melt Area [9] To represent the 2007 Arctic sea ice concentration anomaly Kay et al. [2008] consider the Western Arctic from 120 W to 180 W (180 E to 240 E) and 70 Nto90 N. They claim that in this region a reduced cloudiness and enhanced DSSW radiation contributed to the record 2007 sea ice extent loss. Since the Arctic is highly spatially variable in 3of6

4 Table 2. Summary of the Cloud Fractions Averaged Over June, July and August for Years 2003 to 2007 Broken Down by Region a Cloud Fraction for Each Melt Year 2003 (JJA) 2004 (JJA) 2005 (JJA) 2006 (JJA) 2007 (JJA) 120 E 210 E E 210 E E 240 E a The bold numbers indicate the lowest cloud fractions for 2003 to terms of sea ice concentration changes, we have chosen to divide the Western Arctic into four separate regions to investigate the impact of DSSW on sea ice concentration anomalies. Figure 1 shows the boundaries of each region. Region 1: 120 E to 150 E and 70 Nto90 N; Region 2: 150 E to 180 E and 70 Nto90 N; Region 3: 180 Eto 210 E and 70 Nto90 N; and Region 4: 210 E to 240 E and 70 Nto90 N. Sea ice concentration anomalies for each region during 2003 to 2007 are shown in Figure 2 (solid black line) along with the entire Arctic sea ice concentration anomalies (dashed black line) are given for 0 E E and 70 Nto90 N. Figure 2 clearly shows that regions 1, 2, and 3 capture the 2007 sea ice concentration anomaly, with region 2 showing the most dramatic sea ice decrease. Region 4 shows a small negative sea ice concentration anomaly in To evaluate the relationship between DSSW and the 2007 Arctic Sea ice concentration anomaly it is necessary to include regions 1, 2 as well. Kay et al. [2008] looked at the region between 180 E to 240 E (region 3 and 4). A portion of region 4 does not display characteristics of the large 2007 Arctic Sea ice concentration anomaly and the area does not capture the region with the most dramatic sea ice loss in 2007 (region 2). [10] Kay et al. [2008] use MODIS, CloudSat, and CALIOP observations and ascertain that the region 180 Eto 240 E exhibits large reductions in cloud fraction during the 2007 melt period. Our analysis partially confirms this result; however, closer inspection of the region reveals that only part of the region experienced reduced cloud fraction while another part did not. As previously mentioned, half the region displayed a large decrease in sea ice concentration anomaly for 2007 and the other half showed sea ice concentration anomaly similar to the climatological mean. It is the region with the sea ice concentration similar to the climatological mean which displays the lowest cloud fraction in 2007 in the MODIS record for 2003 to 2007 (Table 2). Figure 3a shows the cloud fraction averaged over June, July, and August 2007, which represents the melt season. Figure 3d shows the monthly averaged sea ice concentration anomaly for September For combined regions 1, 2, and 3 (120 E to 210 E) which fully capture the 2007 Arctic Sea ice concentration anomaly, the lowest cloud fraction occurred in 2005 (Table 2). Part of the region investigated by Kay et al. [2008] for which there is no significant sea ice concentration anomaly in 2007 (210 E to 240 E), has the lowest JJA cloud fraction in While the other part of the region investigated by Kay et al. [2008] (180 E to 210 E) which does display characteristics of the 2007 Figure 3. (a) The spatial distribution of cloud fraction averaged over June, July, and August 2007 (the melt season). (b) The spatial distribution of accumulated DSSW from the beginning of the year until melt onset (June) for (c) The spatial distribution of DSSW radiant exposure accumulated from melt onset (June) to the peak of negative sea ice concentration anomaly (September) for (d) The monthly averaged sea ice concentration anomaly for September of6

5 Figure 4. Time series of the monthly averaged accumulated radiant exposure and the Arctic Sea Ice concentration anomaly. Red line: accumulated radiant exposure; black line: sea ice concentration anomaly. The solid line is for the region 120 E to 210 E while the dashed line is for the region 0 E to 360 E. negative sea ice concentration anomaly, has the lowest JJA cloud fraction in 2005 (Table 2). The fact that the lowest cloud fraction for the JJA melt season of 2007 occurred over areas with positive or no sea ice concentration anomaly seems to indicate that cloud amount was not a primary driver for the 2007 sea ice concentration anomaly Radiant Exposure [11] In order to evaluate the role of various radiative components as drivers of the sea ice anomalies from 2003 to In order to evaluate the role of various radiative components as drivers of the sea ice anomalies from 2003 to 2007 we define the quantity cumulative sum of radiant exposure for each melt year cumulative sum of the radiant exposure for each melt year. Radiant exposure is estimated by multiplying the monthly average value of irradiance by the number of seconds within the month and has the units of Jm 2. For each year between 2003 and 2007 the cumulative sum of radiant exposure was calculated beginning in January of each year. The cumulative sum of the radiant exposure tracks the energy density that an area has received since the beginning of the year, allowing one to compare the amount of energy received from a specific radiation component prior to the beginning of seasonal melt (or specific anomaly) for each year. Using the radiant exposure as a metric for comparison allows a similar starting point for each year investigated DSSW [12] Figure 4 shows the cumulative radiant exposure for DSSW for each year along with the sea ice concentration anomaly for the region 120 E to 210 E and 70 Nto90 N. As noted by others [Kay et al., 2008], there was an increase in DSSW in 2007 as compared to The radiant exposure for 2007 was 2650 MJ m 2, while for 2006 it was 2611 MJ m 2 (Table 3); it also shows that from 2003 to 2007 there were two years with greater radiant exposure than The radiant exposure for 2005 was significantly greater than 2007 with a value of 2850 MJ m 2 and 2004 had a value of 2680 MJ m 2, which was the second highest year in terms of maximum radiant exposure for the period investigated; a positive sea ice concentration anomaly was seen throughout the year. [13] We have also examined the radiant exposure accumulated from the beginning of the year to the melt onset, which is the month of June as shown by Markus et al. [2009]. Again, both 2005 and 2004 exhibit larger radiant exposures as seen in Table 3 and a positive sea ice concentration anomaly exists throughout the year for the second highest radiant exposure accumulated until June. The spatial distribution of accumulated DSSW from the beginning of the year until melt onset for 2007 is shown in Figure 3b. The region of highest accumulated DSSW radiant exposure until melt onset tended to occur over the region where there was no significant negative sea ice concentration anomaly for The regions with lower values of DSSW tend to occur where the 2007 negative sea ice anomaly is significant. The spatial locations of the DSSW are consistent with cloud fraction observations, where higher values of DSSW occur in regions with lower cloud fraction. [14] Figure 3c shows the spatial distribution of DSSW radiant exposure accumulated from melt onset (June) to the peak negative sea ice concentration anomaly (September) for This time interval is selected to determine if the accumulated DSSW during melt season contributed to the anomalous sea ice loss in Figure 3c shows that the area of highest accumulated DSSW occurs over the region where the sea ice concentration anomaly is positive or zero. 4. Conclusions [15] We have examined the role of clouds and DSSW in the Arctic sea ice concentration anomalies between 2003 and 2007; our findings indicate that these factors are not the primary drivers of these anomalies. We observe that the highest values of accumulated DSSW occurred in 2005 consistent with the lowest cloud fractions; however, the largest negative sea ice concentration anomaly occurred in The 2004 season had the second highest value of accumulated DSSW and no negative sea ice concentration anomaly. During the 2007 season, the spatial regions which show the highest amount of negative sea ice concentration anomalies are the same regions which show the lowest values of accumulated DSSW. The inhomogeneity of sea ice melt requires a detailed look on the relationship between irradiance and melt. Table 3. Yearly Values of Accumulated Radiant Exposure and Radiant Exposure Accumulated From the Beginning of the Year to Melt Onset (June) for 2003 to 2007 For Entire Year DSSW Radiant Exposure [MJ m-2] From Beginning of Year to Melt Onset (June) of6

6 [16] Acknowledgments. This work benefited from support under NSF grant ATM and by NOAA grant NA09NES (Cooperative Institute for Climate and Satellites - CICS) at the University of Maryland/ ESSIC. Thanks are due to the NASA GES DISC Giovanni for the MODIS data, to the various MODIS teams that produced data used in this study, to the Baseline Surface Radiation Network and the U.S. Department of Energy Atmospheric Radiation Measurement Climate Research Facility for observations used in evaluation, to the National Snow and Ice Data Center for providing Arctic Sea Ice data and to Xiaolei Niu for her contribution to the development of the MODIS based shortwave radiative fluxes. The Authors thank the anonymous reviewers for very helpful comments. [17] The Editor thanks two anonymous reviewers for assisting in the evaluation of this paper. References Belchansky, G. I., D. C. Douglas, and N. G. Platonov (2004), Duration of the Arctic Sea Ice Melt Season: Regional and Interannual Variability, , J. Clim., 17, 67 80, doi: / (2004) 017<0067:DOTASI>2.0.CO;2. Cavalieri, D., C. Parkinson, P. Gloersen, and H. J. Zwally (1996), Sea ice concentrations from Nimbus-7 SMMR and DMSP SSM/I passive microwave data, digital media, Natl. Snow and Ice Data Cent., Boulder, Colo. Cavalieri, D., C. Parkinson, P. Gloersen, and H. J. Zwally (2008), Sea ice concentrations from Nimbus-7 SMMR and DMSP SSM/I passive microwave data, digital media, Natl. Snow and Ice Data Cent., Boulder, Colo. Comiso, J. (2008), Bootstrap sea ice concentrations from Nimbus-7 SMMR and DMSP SSM/I, , digital media, Natl. Snow and Ice Data Cent., Boulder, Colo. Dong, X., B. Xi, K. Crosby, C. N. Long, R. S. Stone, and M. D. Shupe (2010), A 10 year climatology of Arctic cloud fraction and radiative forcing at Barrow, Alaska, J. Geophys. Res., 115, D17212, doi: / 2009JD Graversen, R. G., T. Mauritsen, S. Drijfhout, M. Tjernstrom, and S. Martensson (2011), Warm winds from the Pacific caused extensive Arctic sea-ice melt in summer 2007, Clim. Dyn., 36, , doi: /s z. Kauker, F., T. Kaminski, M. Karcher, R. Giering, R. Gerdes, and M. Voßbeck (2009), Adjoint analysis of the 2007 all time Arctic sea-ice minimum, Geophys. Res. Lett., 36, L03707, doi: /2008gl Kay, J. E., T. L Ecuyer, A. Gettelman, G. Stephens, and C. O Dell (2008), The contribution of cloud and radiation anomalies to the 2007 Arctic sea ice extent minimum, Geophys. Res. Lett., 35, L08503, doi: / 2008GL Lindsay, R. W., J. Zhang, A. Schweiger, M. Steele, and H. Stern (2009), Arctic sea ice retreat in 2007 follows thinning trend, J. Clim., 22, , doi: /2008jcli Markus, T., J. C. Stroeve, and J. Miller (2009), Recent changes in Arctic sea ice melt onset, freezeup, and melt season length, J. Geophys. Res., 114, C12024, doi: /2009jc Moody, E. G., M. D. King, C. B. Schaaf, D. K. Hall, and S. Platnick (2007), North hemisphere five-year average ( ) spectral albedos of surface in the presence of snow: Statistics computed from Terra MODIS land products, Remote Sens. Environ., 111, , doi: /j.rse Nghiem, S. V., I. G. Rigor, D. K. Perovich, P. Clemente-Colon, J. W. Weatherly, and G. Neumann (2007), Rapid reduction of Arctic perennial sea ice, Geophys. Res. Lett., 34, L19504, doi: /2007gl Niu, X., R. T. Pinker, and M. F. Cronin (2010), Radiative fluxes at high latitudes, Geophys. Res. Lett., 37, L20811, doi: /2010gl Ogi, M., and J. M. Wallace (2007), Summer minimum Arctic sea ice extent and the associated summer atmospheric circulation, Geophys. Res. Lett., 34, L12705, doi: /2007gl Perovich, D. K., B. Light, H. Eicken, K. F. Jones, K. Runciman, and S. V. Nghiem (2007a), Increasing solar heating of the Arctic Ocean and adjacent seas, : Attribution and role in the ice-albedo feedback, Geophys. Res. Lett., 34, L19505, doi: /2007gl Perovich, D. K., S. V. Nghiem, T. Markus, and A. Schweiger (2007b), Seasonal evolution and interannual variability of the local solar energy absorbed by the Arctic sea ice-ocean system, J. Geophys. Res., 112, C03005, doi: /2006jc 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: /2008gl Pinker, R. T., H. Wang, and S. A. Grodsky (2009), How good are ocean buoy observations of radiative fluxes?, Geophys. Res. Lett., 36, L10811, doi: /2009gl Post, E., et al. (2009), Ecological dynamics across the Arctic associated with recent climate change, Science, 325, , doi: / science Rigor, I. G., and J. M. Wallace (2004), Variations in the age of Arctic sea-ice and summer sea-ice extent, Geophys. Res. Lett., 31, L09401, doi: / 2004GL Schweiger, A. J., J. Zhang, R. W. Lindsay, and M. Steele (2008), Did unusually sunny skies help drive the record sea ice minimum of 2007?, Geophys. Res. Lett., 35, L10503, doi: /2008gl Serreze, M. C., M. M. Holland, and J. Stroeve (2007), Perspectives on the Arctic s shrinking sea-ice cover, Science, 315, , doi: /science Shimada, K., T. Kamoshida, M. Itoh, S. Nishino, E. Carmack, F. A. McLaughlin, S. Zimmermann, and A. Proshutinsky (2006), Pacific Ocean inflow: Influence on catastrophic reduction of sea ice cover in the Arctic Ocean, Geophys. Res. Lett., 33, L08605, doi: /2005gl Stroeve, J., M. Serezze, S. Drobot, S. Gearheard, M. Holland, J. Maslanik, W. Meier, and T. Scambos (2008), Arctic sea ice plummets in 2007, Eos Trans. AGU, 89(2), 13, doi: /2008eo Wang, H., and R. T. Pinker (2009), Shortwave radiative fluxes from MODIS: Model development and implementation, J. Geophys. Res., 114, D20201, doi: /2008jd Zhang, T., T. Scambos, T. Haran, L. D. Hinzman, R. G. Barry, and D. L. Kane (2003), Ground-based and satellite-derived measurements of surface albedo on the North Slope of Alaska, J. Hydrometeorol., 4, 77 91, doi: / (2003)004<0077:gbasdm>2.0.co;2. 6of6