DOI: 10.1126/science.276.5311.394. Primary Production in Antarctic Sea Ice

10.1126/science.276.5311.394

Crossref journal-article

American Association for the Advancement of Science (AAAS)

Science (221)

Abstract

A numerical model shows that in Antarctic sea ice, increased flooding in regions with thick snow cover enhances primary production in the infiltration (surface) layer. Productivity in the freeboard (sea level) layer is also determined by sea ice porosity, which varies with temperature. Spatial and temporal variation in snow thickness and the proportion of first-year ice thus determine regional differences in sea ice primary production. Model results show that of the 40 teragrams of carbon produced annually in the Antarctic ice pack, 75 percent was associated with first-year ice and nearly 50 percent was produced in the Weddell Sea.

Bibliography

Arrigo, K. R., Worthen, D. L., Lizotte, M. P., Dixon, P., & Dieckmann, G. (1997). Primary Production in Antarctic Sea Ice. Science, 276(5311), 394â397.

Authors 5

Kevin R. Arrigo (first)
Denise L. Worthen (additional)
Michael P. Lizotte (additional)
Paul Dixon (additional)
Gerhard Dieckmann (additional)

References 22 Referenced 192

H. J. Zwally et al. Antarctic Sea Ice 1973–1976: Satellite Passive Microwave Observations (SP-459 NASA Washington DC 1983).
Arrigo K. R., et al., Mar. Ecol. Prog. Ser. 98, 173 (1993). (10.3354/meps098173) / Mar. Ecol. Prog. Ser. by Arrigo K. R. (1993)
Arrigo K. R., et al., ibid. 127, 255 (1995). / ibid. by Arrigo K. R. (1995)
Stretch J. J., et al., ibid. 44, 131 (1988). / ibid. by Stretch J. J. (1988)
The model includes (i) atmospheric spectral radiation (400 to 700 nm) as a function of time date latitude and cloud cover; (ii) in-ice bio-optics; (iii) sea ice geophysics; and (iv) biological dynamics. Algal production was calculated at each vertical grid point as a function of temperature brine salinity photosynthetically usable radiation and nutrients. Algal accumulation was reduced by a loss term that included the effects of death zooplankton grazing and sinking. Additional details can be found in (6) and (7). The model grid is based on the special sensor microwave/imager (SSM/I) with a horizontal resolution of 625 km 2 . The infiltration and internal freeboard layers were 0.02 m and 0.10 m thick respectively. A uniform ice layer (0.10 m thick) separated the infiltration and freeboard layers. Layer thicknesses were held constant for the length of the integration. Ice thickness below the freeboard layer was variable and was used to distinguish first-year from multiyear ice. First-year ice had a total thickness of 0.75 m whereas the thickest multiyear ice was 1.50 m. The initial distribution of multiyear ice was determined from the minimum ice extent in February 1989. The model was initialized on 1 October 1989 with a uniform chlorophyll a concentration of 1 mg m –3 . During the integration period any multiyear ice that disappeared and then reappeared did so as first-year ice with an initial chlorophyll a concentration of 1 mg m –3 . The model was run with 1-hour time steps at a vertical resolution of 0.5 cm in the infiltration layer and 1 cm in the freeboard layer.
Arrigo K. R., et al., J. Geophys. Res. 98, 6929 (1993). (10.1029/93JC00141) / J. Geophys. Res. by Arrigo K. R. (1993)
Arrigo K. R., Sullivan C. W., Limnol. Oceanogr. 39, 609 (1994). (10.4319/lo.1994.39.3.0609) / Limnol. Oceanogr. by Arrigo K. R. (1994)
In the infiltration layer nutrients were supplied continuously whenever the snow density and thickness were sufficient to force the sea ice below sea level (the ice surface flooded). Nutrients were supplied to the interior freeboard layer when sea ice brine volume a function of temperature and sea ice salinity exceeded 70 per mil (the sea ice became permeable) as well as when the surface flooded. If these criteria were not met the nutrient supply was cut off until the surface flooded or the ice became permeable again.
Legendre L., et al., Polar Biol. 12, 429 (1992). / Polar Biol. by Legendre L. (1992)
The model uses as input the daily sea ice concentrations derived from SSM/I. The model calculates productivity at pixels where sea ice concentration exceeds 50%. Maps of snow distribution were generated from SSM/I imagery (18). Because horizontal variability of snow thickness within the 625-km 2 SSM/I pixel is substantial subpixel variation in snow thickness was simulated by creating nine equal area grid points per pixel. The snow thickness at each grid point was calculated by multiplying the snow depth for that pixel by one of nine multipliers. These multipliers (0.272 0.532 0.952 1.74 3.308 1.305 0.721 0.427 and 0.102) were chosen so that the distribution of snow thickness within each 625-km 2 pixel was consistent with the log-normal frequency distribution observed in situ. The multiplier applied to a given grid point was shifted at most one step to the right or left of the previous multiplier in the array every 4 days to simulate temporal changes in snow thickness at each grid point. The array was ordered such that the difference between adjacent elements was minimized which ensured that temporal changes in snow thickness at each grid point were not too abrupt. For example after a multiplier of 0.721 was applied to a grid point the potential multipliers that could be applied subsequently were 1.305 0.721 or 0.427 with the actual multiplier chosen at random. Production was calculated independently for each grid point and all grid points within a pixel were averaged to obtain the productivity for that pixel. Spatial variation in nutrient concentrations within seawater which supplies the sea ice were obtained from annual climatologies (19). Seawater salinity was obtained from (20). Cloud cover sea-level air pressure relative humidity and sea-level air temperature used in the atmospheric model were obtained from (21). Total column ozone was calculated as the averaged monthly climatologies from years 1989 to 1991 of the total ozone mapping spectrometer data. Total column precipitable water was calculated as the averaged monthly climatologies from years 1985 to 1988 of the TIROS operational vertical sounder data. All monthly climatologies were interpolated daily during the model integration.
Over most of the Antarctic ice pack maximum productivity in the freeboard layer was approximately uniform and close to 1 g C m – 2 month – 1 . However because sea ice in the southwestern Pacific Ocean was constantly disappearing and reforming (on average sea ice at a given location there attained a maximum age of 20 days) rates of production in the freeboard layer were ∼50% lower than in the rest of the ice pack.
Similar proportions of submerged sea ice in the Weddell Sea were reported by P. Wadhams et al. [ J. Geophys. Res. 92 14535 (1987)].
Our estimate of primary production within sea ice must be considered conservative. Pack ice consists of ice of varied age thickness diameter and structural irregularities (for example rafted ice floes snow drifts and refrozen leads) which adds to the potential sites for algal colonization and may result in higher estimates of production.
S. Mathot et al. Eos 76 (no. 3) OS143 (1996).
Smith W. O., Nelson D. M., Bioscience 36, 251 (1986). (10.2307/1310215) / Bioscience by Smith W. O. (1986)
T. K. Frazer L. B. Quetin R. M. Ross in Antarctic Communities: Species Structure and Survival B. Battaglia J. Valencia D. W. H. Walton Eds. (Cambridge Univ. Press Cambridge 1997) pp. 107–111.
Jacobs S. S., Comiso J. C., Geophys. Res. Lett. 20, 1171 (1993). (10.1029/93GL01200) / Geophys. Res. Lett. by Jacobs S. S. (1993)
K. R. Arrigo et al. NASA Tech. Memo. 104640 (1996).
M. E. Conkright et al. World Ocean Atlas 1994 vol. 1 Nutrients (NOAA Atlas NESDIS 1 U.S. Department of Commerce Washington DC 1994).
S. Levitus et al. World Ocean Atlas 1994 vol. 3 Salinity [National Oceanic and Atmospheric Administration (NOAA) Atlas NESDIS 3 U.S. Department of Commerce Washington DC 1994].
A. M. da Silva C. C. Young S. Levitus Atlas of Surface Marine Data 1994 vol. 1 Algorithms and Procedures (NOAA Atlas NESDIS 6 U.S. Department of Commerce Washington DC 1994).
We thank S. Fiegles for assistance with SSM/I imagery; S. Ackley D. Robinson and C. Fritsen for helpful discussions and technical advice; and W. Olson C. McClain W. Esaias L. Harding and A. Schnell for editorial comments. Supported by NSF grant OPP 95-25805 (K.R.A. and M.P.L.) and NASA grants 971-438-20-10 and 971-148-65-56 (K.R.A.).

Dates

Type	When
Created	23 years, 1 month ago (July 27, 2002, 5:45 a.m.)
Deposited	1 year, 7 months ago (Jan. 12, 2024, 11:41 p.m.)
Indexed	1 week, 1 day ago (Aug. 26, 2025, 3:02 a.m.)
Issued	28 years, 4 months ago (April 18, 1997)
Published	28 years, 4 months ago (April 18, 1997)
Published Print	28 years, 4 months ago (April 18, 1997)

Funders 0

None

BibTeX

@article{Arrigo_1997, title={Primary Production in Antarctic Sea Ice}, volume={276}, ISSN={1095-9203}, url={http://dx.doi.org/10.1126/science.276.5311.394}, DOI={10.1126/science.276.5311.394}, number={5311}, journal={Science}, publisher={American Association for the Advancement of Science (AAAS)}, author={Arrigo, Kevin R. and Worthen, Denise L. and Lizotte, Michael P. and Dixon, Paul and Dieckmann, Gerhard}, year={1997}, month=apr, pages={394–397} }

JSON

{
  "indexed": {
    "date-parts": [
      [
        2025,
        8,
        26
      ]
    ],
    "date-time": "2025-08-26T07:02:31Z",
    "timestamp": 1756191751192
  },
  "reference-count": 22,
  "publisher": "American Association for the Advancement of Science (AAAS)",
  "issue": "5311",
  "content-domain": {
    "domain": [],
    "crossmark-restriction": false
  },
  "published-print": {
    "date-parts": [
      [
        1997,
        4,
        18
      ]
    ]
  },
  "abstract": "<jats:p>A numerical model shows that in Antarctic sea ice, increased flooding in regions with thick snow cover enhances primary production in the infiltration (surface) layer. Productivity in the freeboard (sea level) layer is also determined by sea ice porosity, which varies with temperature. Spatial and temporal variation in snow thickness and the proportion of first-year ice thus determine regional differences in sea ice primary production. Model results show that of the 40 teragrams of carbon produced annually in the Antarctic ice pack, 75 percent was associated with first-year ice and nearly 50 percent was produced in the Weddell Sea.</jats:p>",
  "DOI": "10.1126/science.276.5311.394",
  "type": "journal-article",
  "created": {
    "date-parts": [
      [
        2002,
        7,
        27
      ]
    ],
    "date-time": "2002-07-27T09:45:15Z",
    "timestamp": 1027763115000
  },
  "page": "394-397",
  "source": "Crossref",
  "is-referenced-by-count": 192,
  "title": "Primary Production in Antarctic Sea Ice",
  "prefix": "10.1126",
  "volume": "276",
  "author": [
    {
      "given": "Kevin R.",
      "family": "Arrigo",
      "sequence": "first",
      "affiliation": [
        {
          "name": "K. R. Arrigo, NASA Oceans and Ice Branch, Goddard Space Flight Center, Code 971.0, Greenbelt, MD 20771, USA."
        },
        {
          "name": "D. L. Worthen, Science Systems and Applications Inc., Lanham, MD 20706, USA."
        },
        {
          "name": "M. P. Lizotte, Department of Biology and Microbiology, University of Wisconsin\u2013Oshkosh, Oshkosh, WI 54901, USA."
        },
        {
          "name": "P. Dixon, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA 92093, USA."
        },
        {
          "name": "G. Dieckmann, Alfred-Wegener Institut fu\u0308r Polar- und Meeresforschung, Columbusstrasse, D-27570 Bremerhaven, Germany."
        }
      ]
    },
    {
      "given": "Denise L.",
      "family": "Worthen",
      "sequence": "additional",
      "affiliation": [
        {
          "name": "K. R. Arrigo, NASA Oceans and Ice Branch, Goddard Space Flight Center, Code 971.0, Greenbelt, MD 20771, USA."
        },
        {
          "name": "D. L. Worthen, Science Systems and Applications Inc., Lanham, MD 20706, USA."
        },
        {
          "name": "M. P. Lizotte, Department of Biology and Microbiology, University of Wisconsin\u2013Oshkosh, Oshkosh, WI 54901, USA."
        },
        {
          "name": "P. Dixon, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA 92093, USA."
        },
        {
          "name": "G. Dieckmann, Alfred-Wegener Institut fu\u0308r Polar- und Meeresforschung, Columbusstrasse, D-27570 Bremerhaven, Germany."
        }
      ]
    },
    {
      "given": "Michael P.",
      "family": "Lizotte",
      "sequence": "additional",
      "affiliation": [
        {
          "name": "K. R. Arrigo, NASA Oceans and Ice Branch, Goddard Space Flight Center, Code 971.0, Greenbelt, MD 20771, USA."
        },
        {
          "name": "D. L. Worthen, Science Systems and Applications Inc., Lanham, MD 20706, USA."
        },
        {
          "name": "M. P. Lizotte, Department of Biology and Microbiology, University of Wisconsin\u2013Oshkosh, Oshkosh, WI 54901, USA."
        },
        {
          "name": "P. Dixon, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA 92093, USA."
        },
        {
          "name": "G. Dieckmann, Alfred-Wegener Institut fu\u0308r Polar- und Meeresforschung, Columbusstrasse, D-27570 Bremerhaven, Germany."
        }
      ]
    },
    {
      "given": "Paul",
      "family": "Dixon",
      "sequence": "additional",
      "affiliation": [
        {
          "name": "K. R. Arrigo, NASA Oceans and Ice Branch, Goddard Space Flight Center, Code 971.0, Greenbelt, MD 20771, USA."
        },
        {
          "name": "D. L. Worthen, Science Systems and Applications Inc., Lanham, MD 20706, USA."
        },
        {
          "name": "M. P. Lizotte, Department of Biology and Microbiology, University of Wisconsin\u2013Oshkosh, Oshkosh, WI 54901, USA."
        },
        {
          "name": "P. Dixon, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA 92093, USA."
        },
        {
          "name": "G. Dieckmann, Alfred-Wegener Institut fu\u0308r Polar- und Meeresforschung, Columbusstrasse, D-27570 Bremerhaven, Germany."
        }
      ]
    },
    {
      "given": "Gerhard",
      "family": "Dieckmann",
      "sequence": "additional",
      "affiliation": [
        {
          "name": "K. R. Arrigo, NASA Oceans and Ice Branch, Goddard Space Flight Center, Code 971.0, Greenbelt, MD 20771, USA."
        },
        {
          "name": "D. L. Worthen, Science Systems and Applications Inc., Lanham, MD 20706, USA."
        },
        {
          "name": "M. P. Lizotte, Department of Biology and Microbiology, University of Wisconsin\u2013Oshkosh, Oshkosh, WI 54901, USA."
        },
        {
          "name": "P. Dixon, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA 92093, USA."
        },
        {
          "name": "G. Dieckmann, Alfred-Wegener Institut fu\u0308r Polar- und Meeresforschung, Columbusstrasse, D-27570 Bremerhaven, Germany."
        }
      ]
    }
  ],
  "member": "221",
  "reference": [
    {
      "key": "e_1_3_1_2_2",
      "unstructured": "H. J. Zwally et al. Antarctic Sea Ice 1973\u20131976: Satellite Passive Microwave Observations (SP-459 NASA Washington DC 1983)."
    },
    {
      "key": "e_1_3_1_3_2",
      "doi-asserted-by": "crossref",
      "first-page": "173",
      "DOI": "10.3354/meps098173",
      "volume": "98",
      "author": "Arrigo K. R.",
      "year": "1993",
      "unstructured": "Arrigo K. R., et al., Mar. Ecol. Prog. Ser. 98, 173 (1993).",
      "journal-title": "Mar. Ecol. Prog. Ser."
    },
    {
      "key": "e_1_3_1_4_2",
      "first-page": "255",
      "volume": "127",
      "author": "Arrigo K. R.",
      "year": "1995",
      "unstructured": "Arrigo K. R., et al., ibid. 127, 255 (1995).",
      "journal-title": "ibid."
    },
    {
      "key": "e_1_3_1_5_2",
      "first-page": "131",
      "volume": "44",
      "author": "Stretch J. J.",
      "year": "1988",
      "unstructured": "Stretch J. J., et al., ibid. 44, 131 (1988).",
      "journal-title": "ibid."
    },
    {
      "key": "e_1_3_1_6_2",
      "unstructured": "The model includes (i) atmospheric spectral radiation (400 to 700 nm) as a function of time date latitude and cloud cover; (ii) in-ice bio-optics; (iii) sea ice geophysics; and (iv) biological dynamics. Algal production was calculated at each vertical grid point as a function of temperature brine salinity photosynthetically usable radiation and nutrients. Algal accumulation was reduced by a loss term that included the effects of death zooplankton grazing and sinking. Additional details can be found in (6) and (7). The model grid is based on the special sensor microwave/imager (SSM/I) with a horizontal resolution of 625 km 2 . The infiltration and internal freeboard layers were 0.02 m and 0.10 m thick respectively. A uniform ice layer (0.10 m thick) separated the infiltration and freeboard layers. Layer thicknesses were held constant for the length of the integration. Ice thickness below the freeboard layer was variable and was used to distinguish first-year from multiyear ice. First-year ice had a total thickness of 0.75 m whereas the thickest multiyear ice was 1.50 m. The initial distribution of multiyear ice was determined from the minimum ice extent in February 1989. The model was initialized on 1 October 1989 with a uniform chlorophyll a concentration of 1 mg m \u20133 . During the integration period any multiyear ice that disappeared and then reappeared did so as first-year ice with an initial chlorophyll a concentration of 1 mg m \u20133 . The model was run with 1-hour time steps at a vertical resolution of 0.5 cm in the infiltration layer and 1 cm in the freeboard layer."
    },
    {
      "key": "e_1_3_1_7_2",
      "doi-asserted-by": "crossref",
      "first-page": "6929",
      "DOI": "10.1029/93JC00141",
      "volume": "98",
      "author": "Arrigo K. R.",
      "year": "1993",
      "unstructured": "Arrigo K. R., et al., J. Geophys. Res. 98, 6929 (1993).",
      "journal-title": "J. Geophys. Res."
    },
    {
      "key": "e_1_3_1_8_2",
      "doi-asserted-by": "crossref",
      "first-page": "609",
      "DOI": "10.4319/lo.1994.39.3.0609",
      "volume": "39",
      "author": "Arrigo K. R.",
      "year": "1994",
      "unstructured": "Arrigo K. R., Sullivan C. W., Limnol. Oceanogr. 39, 609 (1994).",
      "journal-title": "Limnol. Oceanogr."
    },
    {
      "key": "e_1_3_1_9_2",
      "unstructured": "In the infiltration layer nutrients were supplied continuously whenever the snow density and thickness were sufficient to force the sea ice below sea level (the ice surface flooded). Nutrients were supplied to the interior freeboard layer when sea ice brine volume a function of temperature and sea ice salinity exceeded 70 per mil (the sea ice became permeable) as well as when the surface flooded. If these criteria were not met the nutrient supply was cut off until the surface flooded or the ice became permeable again."
    },
    {
      "key": "e_1_3_1_10_2",
      "first-page": "429",
      "volume": "12",
      "author": "Legendre L.",
      "year": "1992",
      "unstructured": "Legendre L., et al., Polar Biol. 12, 429 (1992).",
      "journal-title": "Polar Biol."
    },
    {
      "key": "e_1_3_1_11_2",
      "unstructured": "The model uses as input the daily sea ice concentrations derived from SSM/I. The model calculates productivity at pixels where sea ice concentration exceeds 50%. Maps of snow distribution were generated from SSM/I imagery (18). Because horizontal variability of snow thickness within the 625-km 2 SSM/I pixel is substantial subpixel variation in snow thickness was simulated by creating nine equal area grid points per pixel. The snow thickness at each grid point was calculated by multiplying the snow depth for that pixel by one of nine multipliers. These multipliers (0.272 0.532 0.952 1.74 3.308 1.305 0.721 0.427 and 0.102) were chosen so that the distribution of snow thickness within each 625-km 2 pixel was consistent with the log-normal frequency distribution observed in situ. The multiplier applied to a given grid point was shifted at most one step to the right or left of the previous multiplier in the array every 4 days to simulate temporal changes in snow thickness at each grid point. The array was ordered such that the difference between adjacent elements was minimized which ensured that temporal changes in snow thickness at each grid point were not too abrupt. For example after a multiplier of 0.721 was applied to a grid point the potential multipliers that could be applied subsequently were 1.305 0.721 or 0.427 with the actual multiplier chosen at random. Production was calculated independently for each grid point and all grid points within a pixel were averaged to obtain the productivity for that pixel. Spatial variation in nutrient concentrations within seawater which supplies the sea ice were obtained from annual climatologies (19). Seawater salinity was obtained from (20). Cloud cover sea-level air pressure relative humidity and sea-level air temperature used in the atmospheric model were obtained from (21). Total column ozone was calculated as the averaged monthly climatologies from years 1989 to 1991 of the total ozone mapping spectrometer data. Total column precipitable water was calculated as the averaged monthly climatologies from years 1985 to 1988 of the TIROS operational vertical sounder data. All monthly climatologies were interpolated daily during the model integration."
    },
    {
      "key": "e_1_3_1_12_2",
      "unstructured": "Over most of the Antarctic ice pack maximum productivity in the freeboard layer was approximately uniform and close to 1 g C m \u2013 2 month \u2013 1 . However because sea ice in the southwestern Pacific Ocean was constantly disappearing and reforming (on average sea ice at a given location there attained a maximum age of 20 days) rates of production in the freeboard layer were \u223c50% lower than in the rest of the ice pack."
    },
    {
      "key": "e_1_3_1_13_2",
      "unstructured": "Similar proportions of submerged sea ice in the Weddell Sea were reported by P. Wadhams et al. [ J. Geophys. Res. 92 14535 (1987)]."
    },
    {
      "key": "e_1_3_1_14_2",
      "unstructured": "Our estimate of primary production within sea ice must be considered conservative. Pack ice consists of ice of varied age thickness diameter and structural irregularities (for example rafted ice floes snow drifts and refrozen leads) which adds to the potential sites for algal colonization and may result in higher estimates of production."
    },
    {
      "key": "e_1_3_1_15_2",
      "unstructured": "S. Mathot et al. Eos 76 (no. 3) OS143 (1996)."
    },
    {
      "key": "e_1_3_1_16_2",
      "doi-asserted-by": "crossref",
      "first-page": "251",
      "DOI": "10.2307/1310215",
      "volume": "36",
      "author": "Smith W. O.",
      "year": "1986",
      "unstructured": "Smith W. O., Nelson D. M., Bioscience 36, 251 (1986).",
      "journal-title": "Bioscience"
    },
    {
      "key": "e_1_3_1_17_2",
      "unstructured": "T. K. Frazer L. B. Quetin R. M. Ross in Antarctic Communities: Species Structure and Survival B. Battaglia J. Valencia D. W. H. Walton Eds. (Cambridge Univ. Press Cambridge 1997) pp. 107\u2013111."
    },
    {
      "key": "e_1_3_1_18_2",
      "doi-asserted-by": "crossref",
      "first-page": "1171",
      "DOI": "10.1029/93GL01200",
      "volume": "20",
      "author": "Jacobs S. S.",
      "year": "1993",
      "unstructured": "Jacobs S. S., Comiso J. C., Geophys. Res. Lett. 20, 1171 (1993).",
      "journal-title": "Geophys. Res. Lett."
    },
    {
      "key": "e_1_3_1_19_2",
      "unstructured": "K. R. Arrigo et al. NASA Tech. Memo. 104640 (1996)."
    },
    {
      "key": "e_1_3_1_20_2",
      "unstructured": "M. E. Conkright et al. World Ocean Atlas 1994 vol. 1 Nutrients (NOAA Atlas NESDIS 1 U.S. Department of Commerce Washington DC 1994)."
    },
    {
      "key": "e_1_3_1_21_2",
      "unstructured": "S. Levitus et al. World Ocean Atlas 1994 vol. 3 Salinity [National Oceanic and Atmospheric Administration (NOAA) Atlas NESDIS 3 U.S. Department of Commerce Washington DC 1994]."
    },
    {
      "key": "e_1_3_1_22_2",
      "unstructured": "A. M. da Silva C. C. Young S. Levitus Atlas of Surface Marine Data 1994 vol. 1 Algorithms and Procedures (NOAA Atlas NESDIS 6 U.S. Department of Commerce Washington DC 1994)."
    },
    {
      "key": "e_1_3_1_23_2",
      "unstructured": "We thank S. Fiegles for assistance with SSM/I imagery; S. Ackley D. Robinson and C. Fritsen for helpful discussions and technical advice; and W. Olson C. McClain W. Esaias L. Harding and A. Schnell for editorial comments. Supported by NSF grant OPP 95-25805 (K.R.A. and M.P.L.) and NASA grants 971-438-20-10 and 971-148-65-56 (K.R.A.)."
    }
  ],
  "container-title": "Science",
  "original-title": [],
  "language": "en",
  "link": [
    {
      "URL": "https://www.science.org/doi/pdf/10.1126/science.276.5311.394",
      "content-type": "unspecified",
      "content-version": "vor",
      "intended-application": "similarity-checking"
    }
  ],
  "deposited": {
    "date-parts": [
      [
        2024,
        1,
        13
      ]
    ],
    "date-time": "2024-01-13T04:41:30Z",
    "timestamp": 1705120890000
  },
  "score": 1,
  "resource": {
    "primary": {
      "URL": "https://www.science.org/doi/10.1126/science.276.5311.394"
    }
  },
  "subtitle": [],
  "short-title": [],
  "issued": {
    "date-parts": [
      [
        1997,
        4,
        18
      ]
    ]
  },
  "references-count": 22,
  "journal-issue": {
    "issue": "5311",
    "published-print": {
      "date-parts": [
        [
          1997,
          4,
          18
        ]
      ]
    }
  },
  "alternative-id": [
    "10.1126/science.276.5311.394"
  ],
  "URL": "http://dx.doi.org/10.1126/science.276.5311.394",
  "relation": {},
  "ISSN": [
    "0036-8075",
    "1095-9203"
  ],
  "subject": [],
  "container-title-short": "Science",
  "published": {
    "date-parts": [
      [
        1997,
        4,
        18
      ]
    ]
  }
}