Research ArticleCLIMATOLOGY

Rapid ablation zone expansion amplifies north Greenland mass loss

See allHide authors and affiliations

Science Advances  04 Sep 2019:
Vol. 5, no. 9, eaaw0123
DOI: 10.1126/sciadv.aaw0123
  • Fig. 1 High-resolution runoff patterns and post-1990 changes.

    (A) Annual mean runoff (RU) over the GrIS and peripheral glaciers and ice caps (1958–2017) as modeled by RACMO2.3p2 and statistically downscaled to 1-km horizontal resolution. Yellow stars locate 213 stake sites where SMB is measured. The seven selected sectors of the GrIS derived from (16) are overlaid. (B) Time series of annual cumulative runoff integrated over the GrIS (red). Panels (C), (D), and (E) show the same for the SW, NW, and NO sectors, respectively, of the GrIS [see (A)]. Dashed red lines show linear runoff trends over 1991–2017, and dashed black lines show the averaged runoff over the periods 1958–1990 and 1991–2017, respectively. Dashed yellow lines mark exceptional runoff years, i.e., 3 SDs (σ) above the 1958–1990 mean. Dashed gray lines highlight recent exceptional runoff years over the GrIS, i.e., 2010, 2012, and 2016 (see fig. S7). Individual trends (1991–2017), relative increase in runoff post-1990 (%), and SD (σ) for the period 1958–1990 are listed at the bottom of each subpanel.

  • Fig. 2 Regional changes in runoff contribution.

    (A) Average contribution of individual GrIS sectors to runoff totals for the period 1958–1990. (B) Post-1990 change in runoff contribution per sector (1991–2017 minus 1958–1990). Sectors experiencing significant change in runoff contribution, i.e., based on Student’s t test (t ≤ 0.10), are stippled with white dots. The post-1990 relative change in runoff contribution is also listed for each sector in white. (C) Time series of runoff contribution for North (blue; i.e., NO + NW) and South (red; i.e., CW + SW) Greenland. Post-1990 trends in runoff contribution are displayed as dashed lines.

  • Fig. 3 Rapid ablation zone expansion enhances runoff contribution from north Greenland.

    (A) Map of SMB averaged for the period 1958–1990. Numbers refer to the ablation zone area for individual sectors (103 km2) and for the whole GrIS (bottom right). (B) Same as (A) for the period 1991–2017. Numbers refer to the relative ablation zone expansion (%) post-1990 for individual sectors and for the whole GrIS (bottom right). Time series of annual mean modeled ablation zone and summer bare ice area for (C) North Greenland (blue and cyan; i.e., NW + NO) and (D) South Greenland (red and orange; i.e., CW + SW sectors) over the period 1958–2017 compared to MODIS (black) bare ice area estimates (2000–2018). Dashed lines show the 1991–2017 trends. Numbers include trends, relative ablation/bare ice zone expansion, and change in ELA (i.e., SMB = 0) post-1990. In North and South Greenland, the modeled bare ice area is averaged over 10 and 5 days, respectively (see Materials and Methods). The cyan and yellow belt in (C) and (D) represents 1 SD of the 10 and 5 days used to estimate the modeled maximum bare ice area in North and South Greenland, respectively. The gray belt in (C) and (D) shows the uncertainty in measured MODIS bare ice area (see Eq. 1 in Materials and Methods).

  • Fig. 4 Recent shift in summer atmospheric circulation and impact on cloudiness.

    (A) Post-1990 change in summer cloud content (JJA; 1991–2017 minus 1958–1990) as modeled by RACMO2.3p2 at 5.5 km. Changes in large-scale circulation (black vectors; see inset for wind speed estimation) and in height of the 500-hPa geopotential (dashed black lines) are overlaid. (B) Change in modeled SWd and (C) LWd radiation. In (C), sector-averaged near-surface temperature (2 m) increase is displayed in black and white for southern and northern Greenland, respectively. Average GrIS-wide temperature increase at 500 hPa and at 2-m altitude is listed in the inset. The seven GrIS sectors are outlined in gray in (A) to (C). Stipples highlight regions showing a significant change, i.e., > 1 SD of the 1958-1990 period (σ) in (A) cloud content (σ = 4 g m−2), (B) SWd (σ = 1.9 W m−2), and (C) LWd (σ = 1.7 W m−2).

  • Fig. 5 Increased early-summer cloudiness triggers runoff amplification in north Greenland.

    (A) Time series of daily (April-September) mean runoff contribution to GrIS totals for periods 1958–1990 and 1991–2017 in South (orange and red; i.e., CW + SW sectors) and North (cyan and blue; i.e., NO + NW sectors) regions. (B) Time series of daily (April-September) cumulative anomalies (1991–2017 minus 1958–1990) in surface melt (orange) runoff (red) for the North region (NO + NW sectors; right y axis). Dashed gray and cyan lines (left y axis) show cumulative anomalies in cloud content and refreezing capacity, i.e., the fraction of meltwater and rainwater retained and/or refrozen in the firn. Anomalies in (C) June LWn and (D) July-August SWn. (E) Daily exposed bare ice area post-1990 and pre-1991 for the South (red and orange) and North (blue and cyan) regions expressed as a fraction (%) of the ablation zone area. Numbers at the bottom left corner of (E) express the relative increase (%) in maximum bare ice area post-1990 for North (blue) and South (red) Greenland. In (A), (B), and (E), the gray and yellow shades outline the period during which runoff contribution of North Greenland significantly increases (A) under high and low cloudiness successively (B).

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/5/9/eaaw0123/DC1

    Table S1. Changes in runoff production and contribution per sector.

    Fig. S1. RACMO2.3p2 integration domain and SMB evaluation.

    Fig. S2. Evaluation of modeled meteorological variables at 5.5 km.

    Fig. S3. Evaluation of radiative fluxes at 5.5 km.

    Fig. S4. Evaluation of the downscaled product using in situ and catchment measurements.

    Fig. S5. Evaluation of the modeled bare ice area using remote sensing.

    Fig. S6. Post-1990 upward migration of the equilibrium line.

    Fig. S7. High interannual variability in summer atmospheric circulation and impacts on the cloudiness.

    Fig. S8. Reduced summer cloud cover enhances melt in southern Greenland.

    Fig. S9. Post-1990 changes in surface conditions.

  • Supplementary Materials

    This PDF file includes:

    • Table S1. Changes in runoff production and contribution per sector.
    • Fig. S1. RACMO2.3p2 integration domain and SMB evaluation.
    • Fig. S2. Evaluation of modeled meteorological variables at 5.5 km.
    • Fig. S3. Evaluation of radiative fluxes at 5.5 km.
    • Fig. S4. Evaluation of the downscaled product using in situ and catchment measurements.
    • Fig. S5. Evaluation of the modeled bare ice area using remote sensing.
    • Fig. S6. Post-1990 upward migration of the equilibrium line.
    • Fig. S7. High interannual variability in summer atmospheric circulation and impacts on the cloudiness.
    • Fig. S8. Reduced summer cloud cover enhances melt in southern Greenland.
    • Fig. S9. Post-1990 changes in surface conditions.

    Download PDF

    Files in this Data Supplement:

Navigate This Article