Greenland ice albedo decline

A new study by Polashenski and others find that “Neither dust nor black carbon causing apparent albedo decline in Greenland’s dry snow zone; implications for MODIS C5 surface reflectance“.

Just one point of their work I expand on here is the issue of declining MODIS Terra sensor sensitivity (Wang and others 2012; Lyapustin et al. (2014). First, Polashenski and others (2015) develop the MODIS albedo decline issue nicely. Now, rewinding back to early 2012, after becoming aware of Wang and others (2012) results, I evaluated whether the albedo decline was present in the completely independent Greenland Climate Network (GC-Net) data after Steffen and others (1996). In 2012, I wrote:

“Degrading MODIS instrument sensitivity identified by Wang et al. (2012) introduces the possibility that the declining albedo trends may be erroneous. To validate the MODIS albedo trends, coinciding observations from GC-Net AWS are examined. The ground truth data are situated across a range of elevations, spanning the ablation and accumulation areas. Analysis of the GC-Net data confirms declining albedo trends in the 2000–2010 period to be widespread in individual months from May–September. Trend statistics are computed where at least 7 yr of annual data are available from both GC-Net and MODIS Terra. Significance is designated here more strictly where the trend measured by the linear regression slope has a magnitude that exceeds 2 of the residuals from the regression. In 41 of 43 (95 %) of monthly cases May–September, the trend is found to be significant and decreasing (Table 1). In 10 of 14 (71 %) cases, for which both GC-Net and MOD10A1 trends are significant, the GC-Net declining trend is larger than the MOD10A1 trend. It therefore does not seem that MODIS sensor degradation is enhancing an existing trend.” – Box and others (2012) 

Still, an update after 2010 is in order. Now, the evaluation through 2014 yields that there is still a real albedo decline for the southern part of the ice sheet, including places like Saddle or South Dome where surface melting is uncommon. Note how not only do both GC-Net and MODIS MOD10A1 show a decline, they share peaks and troughs. Given that the ground data having a footprint size of just a few square meters and the satellite data that have an effective footprint size of 5 x 5 km and that they pick up the same high and low years is impressive.


Wonk Alert: Still the bias is very likely a latitude-dependent. Notice how the GC-Net trend is even stronger than the MODIS trend in the far south (South Dome and Saddle). Consider that we have sunlight reflecting off of a highly reflective part (an ice sheet) of a sphereoid (the earth) that arcs more than 20 degrees north south. Snow and ice have reflectance depending strongly on viewing and illumination angles. So, the satellite data compensation for simple albedo are susceptible to amplification of small sensor degradation biases.
Let me add that the detected bias is smaller than the type of anomalies produced by for example the large July 2015 melting for NW Greenland. The red and yellow areas below are real local albedo anomalies due primarily to melting.

Alb_LA_EN_20150709 NW Greenland stays melty is widespread west and expands higher in elevation

MODIS having a more negative trend than the ground data for the northern high and usually dry bright snow interior sites (See NASA-E, Tunu-N, Humboldt above) is also a real issue that Polashenski and others (2015) nicely report. While the bias we are looking forward to be compensated in the version 6 MODIS data (Lyapustin and othes 2014). As to the role of black carbon in Greenland’s albedo decline, I would say there is more to the story than what Polashenski and others (2015) report. Stay tuned.

Works Cited

  • Box, J. E., X. Fettweis, J.C. Stroeve, M. Tedesco, D.K. Hall, and K. Steffen. 2012. Greenland ice sheet albedo feedback: thermodynamics and atmospheric drivers, The Cryosphere, 6, 821-839. doi:10.5194/tc-6-821-2012
  • Lyapustin, A., Wang, Y., Xiong, X., Meister, G., Platnick, S., Levy, R., Franz, B., Korkin, S., Hilker, T., Tucker, J., Hall, F., Sellers, P., Wu, A., and Angal, A.: Scientific impact of MODIS C5 calibration degradation and C6+ improvements, Atmos. Meas. Tech., 7, 4353-4365, doi:10.5194/amt-7-4353-2014, 2014.
  • Polashenski, C.M. J.E. Dibb, M.G. Flanner, J.Y. Chen, Z.R. Courville, A.M. Lai, J.J. Schauer, M.M. Shafer and M. Bergin, 2015, Neither dust nor black carbon causing apparent albedo decline in Greenland’s dry snow zone; implications for MODIS C5 surface reflectance, DOI: 10.1002/2015GL065912
  • Steffen, K., Box, J. E., and AbdalatiW.: Greenland climate network: GCNet, US Army Cold Regions Reattach and Engineering (CRREL), CRREL Special Report, 98–103, 1996.
  • Wang, D. D., Morton, D., Masek, J., Wu, A. A., Nagol, J., Xiong, X., Levy, R., Vermote, E., and Wolfe, R.: Impact of sensor degradation on the MODIS NDVI time series, Remote Sens. Environ., 119, 55­61, 2012.


record setting 2015 North American fire radiative power – 2.5x the 2000-2015 average

Sensors in Earth orbit give us the capability to monitor vast areas, daily, in near real-time. I’ve been working with daily NASA MODIS MOD14A1 data to map seasonal fire activity since the data begin year 2000. The map below illustrates the single most active day so far in 2015 for North America with fires ravaging central western Canada and interior Alaska.


A new strongest fire season?

Through 18 July, 2015, these data indicate the cumulative radiative power of North American fires to be the highest on record in the period of observations beginning in 2000. For July, 2015 fire power is 2.5 times the sixteen summer average 2000-2015. The fire season spikes above the annual average earlier in the year than in other years.


reporting from [29 June, 2015]:

Western Canada experienced more than 600 fires over the weekend, according to territorial authorities.

Canadian provinces and territories pool their firefighting resources in these circumstances. While the NWT has requested more backup, other provinces are perceived to have a “dire need” and are first in line.

“Saskatchewan, for instance, is undergoing a series of evacuations of communities,” said Frank Lepine, the territory’s associate director of forest management. “Manitoba is pretty close to that.

“The NWT will be receiving some single resources but no more crews at this time. [But] that may change by the end of the week.”

There are 129 fires burning in the NWT, which has experienced a total of 158 fires so far this season. The 20-year average is 66 fires for this time of year.

And the 2015 fire season is not yet over.

According to this analysis, the previous  year 2014 ended by setting the annual record for cumulative fire power for North America. Year 2004 fires were concentrated around the Alaska Canada border.

Scanning past news stirs recent intense memories….

High Country News [2014]:

Canada’s Northwest Territories are on fire. The region is experiencing its hottest, driest summer in 50 years, and wildfire activity is more than six times the 25-year average. While blazes in sparsely populated northern Canada have a minimal impact on human safety and infrastructure, they have an outsized effect on the environment: The ancient, stunted boreal forests, or taiga, ringing the Arctic Circle contain 30 percent of the world’s land-based carbon.


July 2, 2004 — A pall of smoke the size of Texas continues to blanket most of Alaska, as several dozen wildfires continue to burn out of control. More than a million acres have burned in the state. There are currently 61 active fires in the state, mostly in the eastern interior

NOAA News [December, 2004]:

Alaska had a record warm summer with a statewide temperature of 4.6 degrees F (2.6 degrees C) above the 1971-2000 mean. May, June, July and August were all record breaking for the state.

Is the smoke drifting to Greenland?

Dark Snow Project has been busy gathering field data this summer in Greenland. We shall report on these. So, stay tuned!

Follow us on Twitter and Facebook and consider pledging support as we work on a tight budget!

fire, ice, soot, carbon: Dark Snow Project 2014 final field work in Greenland

Arrived yesterday to Kangerlussuaq, west Greenland, now 6 AM, we’re just about out the door in effort to put more numbers on how fire and other factors are affecting Greenland’s reflectivity as part of the Dark Snow Project.

I just received this 27 July, 2014 NASA MODIS satellite image showing wildfire smoke drifting over Greenland ice.

Premier climate video blogger Peter Sinclair is a key component of the Dark Snow Project because of our focus on communicating our science to the global audience. The video below was shot and edited last night quickly as we prepare for a return to our camp a few hours from now.

The video does not comment on the important issue of carbon. So, here’s a quick research wrap-up… Wildfire is a source of carbon dioxide, methane and black carbon to the atmosphere. Jacobson (2014) find that sourcing to be underestimated in earlier work. Graven et al. (2013) find northern forests absorbing and releasing more carbon by respiration due to Arctic warming’s effects on forest composition change. At the global scale, the land environment produces a net sink of carbon, taking up some 10% of the atmospheric carbon emissions due to fossil fuel combustion (IPCC, 2007). Yet, whether northern wildfire is becoming an important source of atmospheric carbon (whether from CO2 or CH4 methane) remains under investigation. University of Wisconsin-Madison researchers find:

“fires shift the carbon balance in multiple ways. Burning organic matter quickly releases large amounts of carbon dioxide. After a fire, loss of the forest canopy can allow more sun to reach and warm the ground, which may speed decomposition and carbon dioxide emission from the soil. If the soil warms enough to melt underlying permafrost, even more stored carbon may be unleashed.

“Historically, scientists believe the boreal forest has acted as a carbon sink, absorbing more atmospheric carbon dioxide than it releases, Gower says. Their model now suggests that, over recent decades, the forest has become a smaller sink and may actually be shifting toward becoming a carbon source.

“The soil is the major source, the plants are the major sink, and how those two interplay over the life of a stand really determines whether the boreal forest is a sink or a source of carbon

Works Cited
  • Danish Meterological Institute provided the NASA MODIS satellite image
  • Graven, H.D., R. F. Keeling, S. C. Piper, P. K. Patra, B. B. Stephens, S. C. Wofsy, L. R. Welp, C. Sweeney, P.P. Tans, J.J. Kelley, B.C. Daube, E.A. Kort, G.W. Santoni, J.D. Bent, 2013, Enhanced Seasonal Exchange of CO2 by Northern Ecosystems Since 1960,  Science: Vol. 341 no. 6150 pp. 1085-1089, DOI: 10.1126/science.1239207
  • Climate Change 2007: Working Group I: The Physical Science Basis, IPCC Fourth Assessment Report: Climate Change 2007
  • Jacobson, M. Z., 2014, Effects of biomass burning on climate, accounting for heat and moisture fluxes, black and brown carbon, and cloud absorption effects, J. Geophys. Res. Atmos., 119, doi:10.1002/2014JD021861.

2014 Greenland ice sheet reflectivity near record low

The NASA MODIS sensor on the Terra satellite provides surface reflectivity data since early 2000 enabling us to evaluate just how dark Greenland ice is today and in comparison with the past 14 years.

The data show that 2014 ice sheet reflectivity (also called albedo) has been near record low much of 2014, especially at the highest elevations.


15 years of albedo data for the uppermost region of the ice sheet

The darkness of the surface at high elevations is consistent with the findings of Dumont et al. (2014) that an increasing dust concentration on the ice sheet in the pre-melt season from decreasing snow cover on land upwind of the ice sheet may be a significant darkening factor.

If there will be a persistent pattern of warm air brought over the ice sheet as in 2012, we should expect melting at the ice sheet upper elevations. Why? Low reflectivity heats the snow more than normal, removing more of the ‘cold content’. A dark snow cover will thus melt earlier and more intensely. A positive feedback exists for snow in which once melting begins, the surface gets yet darker due to increased liquid water content, increased snow grain size, and possible other factors such as microbial growth.

For the ice sheet as a whole, low reflectivity in 2014 has been exceeded only by years 2012, 2013, and 2011, depending on the time of year…


15 years of albedo data for the entire ice sheet and peripheral glaciers

The Greenland reflectivity anomaly map features red and orange colors that indicate a relatively dark surface near the end of June especially at the low elevations where most melting occurs.

albedo anomaly map

albedo anomaly map. For more, see

Work Cited

  1. Dumont, M., E. Brun, G. Picard, M. Michou, Q. Libois, J-R. Petit, M. Geyer, S. Morin and B. Josse, Contribution of light-absorbing impurities in snow to Greenland’s darkening since 2009, Nature Geoscience, 8 June, 2014, DOI: 10.1038/NGEO2180

west Greenland melt is ON

Weather was very warm yesterday in Kangerlussuaq, at least 15 C (60 F) but 20 C (70 F) at the unofficial airport site. The river came up fast and wide between our 10 AM first look to the late afternoon; 24 h sun here on the Arctic circle. The snowline is migrating up the ice sheet. It seems we arrived right on the start of continuous melt. The previous days have had variable weather and even fresh snow on hilltops.

The melt is ON. The extended forecast is for warm sunny weather…

Screen Shot 2014-06-11 at 8.23.18 AMThe warm weather is a relief because right now, snowline is ~850 m above sea level. We aim to camp at 1250 m and don’t want to arrive to slush deeper than our ankles.

The precipitation forecast for next Wednesday would come the night of our camp put in. We’d rather have snow than rain. But the freezing level in the atmosphere would be right at camp elevation, so it would be a ‘wintery mix’.