TG-1 * Transgallaxys Forum 1

Advanced search  
Pages: [1]

Author Topic: Mother Jones gets ridiculous  (Read 270 times)


  • Jr. Member
  • *
  • Posts: 50
Mother Jones gets ridiculous
« on: September 30, 2019, 10:17:42 AM »

Read and wonder:


September 26, 2019
Arctic Ice Is Melting Faster Than Expected. These Scientists Have a Radical Idea to Save It.
Costly and controversial, yes. But we may be running out of choices.

    Marisa Endicott

Krista Kahl/Ice911

Tiny glass beads might seem an unlikely hero in the fight against climate change, but they may end up playing an outsize role in tackling one of the natural world’s most dire predicaments. A group of researchers have found that millions of these spheres spread in a layer across swaths of Arctic ice reflects sunlight and helps keep the ice frozen.
During a heatwave this summer, Greenland experienced the biggest loss of ice in a 24-hour period since record keeping began.

Glaciers, polar land, and sea ice are rapidly melting, much faster than many scientists expected, as the Intergovernmental Panel on Climate Change’s (IPCC) report on oceans and the cryosphere released Wednesday reiterated. During a five-day heatwave this summer, Greenland lost more than 60 billion tons of ice, including the biggest loss in a 24-hour period since record-keeping began. Recent projections warn that Arctic summers could be nearly ice-free in 10 to 40 years. Arctic ice levels are both a primary indicator of rising global temperatures and a key determinant of how bad climate change might get. Serious ice loss means sea levels rise, more heat is absorbed by the ocean and planet, and weather and ocean currents could change.
Advertise with Mother Jones

These grim warnings have prompted more researchers to apply technological solutions that intervene in the earth’s climate systems to slow the impacts of global warming, also known as “geoengineering.” Last week, leading up to the United Nations Climate Action Summit on Monday, a contingent of climate organizations partnered with the UN to host the First Annual Global Climate Restoration Forum to discuss geoengineering solutions. While climate engineering was previously considered too far-fetched, risky, or controversial, more people are now seeing the strategy as a necessary piece of the puzzle.

“We’re getting much more attention because of how dire thing are,” says inventor and engineer Leslie Field, founder of Ice911, a nonprofit exploring arctic ice restoration who featured prominently at the climate restoration conference and in the strategic report organizers released.

“The ice melting is a critical game changer that we need to stop as fast as possible,” says Rick Parnell, CEO of the Foundation for Climate Restoration, the organization behind the forum at the UN. One 2018 report concluded that geoengineering polar glaciers would slow sea level rise and buy significant extra time to tackle climate change. “Stopping arctic melt is key,” Field says. “This shouldn’t be an afterthought.”

The Arctic loses significant ice cover in the spring and summer.


In 2006, Field saw Al Gore’s climate documentary An Inconvenient Truth. She couldn’t stop thinking about her two young children and the future they faced. But rather than despair, she decided to get to work. Field calls her idea “an embarrassingly simple concept.” Warmer temperatures have reduced the bright white multiyear ice (essentially thicker, stiffer ice) in the Arctic, and the resulting loss in light reflectivity leads to heating and further ice melt. “I just asked myself a very simple question: Is there a safe material that could help replace that lost reflectivity?” Field says. The answer, her team found, was to add a very thin layer of hollow glass spheres made of silica, the main component of most rocks, that reflect light and “make very thin ice look a lot more like multi-year ice.”,975

            Sign up for our newsletters

Subscribe and we'll send Mother Jones straight to your inbox.

The silica beads resemble fine sand or salt and are made from the main component in most rocks.


Ice911 isn’t the only ice restoration project out there. People have proposed building underwater walls to prop up glaciers and boring cold water tunnels under ice to help thicken it. Chinese researchers are investing billions into polar research over the next decade, including feasibility testing for various geoengineering solutions.

But Ice911 is one of the furthest along. Last year, the team covered almost 18,000 square yards on lake ice in Utqiagvik (previously Barrow), Alaska, for pilot tests. While they work on permitting with the Environmental Protection Agency for testing on sea ice, they are collaborating with NASA on modeling.

The results so far are encouraging. In May 2018, the team published a peer-reviewed paper in the American Geophysical Union’s Earth’s Future journal, revealing the results and projections of their initial pilot testing. In one field test, the material increased reflectivity by 15 to 20 percent. Their models have predicted the beads could significantly stop ice decline and even reverse it. Simulations showed a potential 1.5 degrees Celsius temperature reduction over a large part of the Arctic, a 3-degree sea temperature reduction in some areas, and an increase in sea ice thickness of up to 20 inches.

Compared to other ice restoration ideas, Ice911’s approach is relatively small-scale. Rather than try to blanket the entire Arctic, the team wants to target specific areas that modeling determines will have the biggest restorative impact. Led by a guiding principle of “do no harm,” Field says “the whole idea is minimal intervention for maximum good.” Safety testing of the silica beads by the manufacturer and an independent lab has yet to turn up any negative effects for humans or animals.
Advertise with Mother Jones

The Ice911 team has field tested its silica microspheres in Alaska.


That doesn’t mean the project is without criticism. Adrienne Titus, an Inupiaq organizer in Fairbanks, Alaska, has raised concerns about the unknown environmental consequences of spreading glass beads all over the ice. Arctic explorers and oil drillers—who historically rarely consulted native communities—have made many indigenous groups wary of intervention. That’s part of why Field is currently focused on engaging governments and native communities, trying to spur conversations about the necessary governance policies and oversight required before any such project can be deployed at a large scale. Ultimately, Field argues, “it’s an even bigger change to the ecosystem to allow the ice melt to continue and not do anything about it, then it would be to make a mild intervention.”

Some climate scientists and policymakers feel the financial resources required to execute these projects at scale—Ice911 estimates its project would cost up to an estimated $5 billion, and other projects would cost much more—would be better used on tried-and-true methods of climate change mitigation.

“A much more efficient way of using funding nowadays would be to subsidize electric cars and solar panels and other alternative energy sources [and] research on electricity storage,” says Slawomir Tulaczyk, a glaciologist at the University of California–Santa Cruz, who has been working in the field for over two decades. While he is personally intrigued by ice engineering work, “these would be probably more realistic investments that are more likely to result in the outcome you want than what would be associated with going to Greenland and Antarctica just to test,” he says.
“As far as we can tell, this is the largest single lever that one could safely address on climate change right now.”

Perhaps most indicative of the tricky space in which climate restoration exists is the field’s persistent moral hazard problem: Some environmentalists fear that people will ignore the causes of climate change if they think they can engineer their way out of it. Sen. Bernie Sanders has called geoengineering a “false solution.” Indeed, fossil fuel companies have ducked addressing climate change by framing it as simply an engineering issue, as former Secretary of State and Exxon CEO Rex Tillerson has repeatedly claimed.


            Get our award-winning magazine

Save big on a full year of investigations, ideas, and insights.

Field is familiar with this argument. It was her main hesitation in publicizing her project at first. But as climate disaster looms, she and others see a pressing need to attack the problem from all angles. Even international organizations like the UN and the IPCC have been signaling more openness to climate engineering ideas, for instance by participating in the climate restoration conference and referencing the potential usefulness of such methods in reports.

“We never have intended nor could we replace the need to do emissions reduction and get sustainable energy sources in place,” Field says. “This is part of a much larger picture that needs to happen, but it’s an essential part. As far as we can tell, this is the largest single lever that one could safely address on climate change right now.”
More MotherJones reporting on Climate Desk

The weeks and months ahead will be among the most consequential of our political lives: a president who puts his interests above the public's; who flouts the law; and who, up to this point, has been protected by an administration and legislators who have looked the other way.

Mother Jones—and that includes you, our community of readers and changemakers—was built for a moment like this, and we're going into overdrive to separate fact from fiction.

Follow the latest impeachment developments here, and please consider supporting our journalism with a donation today. Your contribution will be matched, dollar for dollar, up to $500,000 total, as part of our special campaign, The Moment for Mother Jones.

    More about:Climate ChangeClimate DeskEnvironmentEnvironmental Protection Agency


    There Is an Important Climate Solution Democrats Never Talk About. But Why?

    Maddie Stone
    Antarctic Sea Ice Is “Astonishingly” Low This Melt Season

    Eric Holthaus
    Solutions to Climate Change Cannot Overlook This One Key Factor

    Alice C. Hill
    Some Indigenous Communities Have a New Way to Fight Climate Change: Give Personhood Rights to Nature

    Anna V. Smith

We Recommend

    Why the Labor Strike Is Back

    Bryan Schatz
    The Beltway Sniper Is Now the Center of a Debate About Juvenile Lifers

    Samantha Michaels
    The Biden Campaign Is Demanding That TV Execs Stop Booking Guiliani

    Ali Breland
    The Washington Post Published a Bombshell Report About Trump and Russia

    Tim Murphy


    Donald Trump Suggests Adam Schiff Should Be Arrested for “Treason”

    Inae Oh
    Online Content Depicting Child Sexual Abuse Is Growing at an Alarming Rate

    Ali Breland
    A New Poll Shows a Majority of Americans Approve of Trump Impeachment Inquiry

    Ali Breland
    Stephen Miller Calls the Whistleblower a “Deep State” Agent

    Ali Breland

We have a new comment system! We are now using Coral, from Vox Media, for comments on all new articles. We'd love your feedback.
view comments
Independent. In print. In your mailbox.

Inexpensive, too! Subscribe today and get a full year of Mother Jones for just $12.
Mother Jones Magazine Cover : September + October 2019 October 2019
Mother Jones Magazine Cover : July + August 2019 August 2019
Mother Jones Magazine Cover : May + June 2019 June 2019
Smart. Fearless. Audio.

It's us but for your ears. Listen on Apple Podcasts.
Subscribe »
Looking for news you can trust?

Subscribe to our free newsletters.

    Our Staff
    Contact Us

    Ways to Give
    Advertise With Us

    Magazine Subscriptions
    Give a Gift Subscription
    Email Newsletters
    Customer Service

Copyright © 2019 Mother Jones and the Foundation for National Progress. All Rights Reserved.

    Terms of Service Privacy Policy

Powered by VIP

Ever heard of a snowstorm?  Then think about the tiny particles being whirled through the air, all over the planet, because the wind does not stop.

They want to cover the whole surface of earth with white dust particles. How many meters high should this white layer be?

Whatever we do, it must not be blown away by the wind!


  • Jr. Member
  • *
  • Posts: 50
Re: Mother Jones gets ridiculous
« Reply #1 on: September 30, 2019, 10:20:29 AM »

Their web designer should be fired.

AGU Journals

    Other Publications


Earth's FutureVolume 6, Issue 6
Research Article Open Access
Increasing Arctic Sea Ice Albedo Using Localized Reversible Geoengineering
L. Field
D. Ivanova
S. Bhattacharyya
V. Mlaker
A. Sholtz
R. Decca
A. Manzara
D. Johnson
E. Christodoulou
P. Walter
K. Katuri
First published: 21 May 2018
This article was corrected on 28 AUG 2018. See the end of the full text for details.



    1 Introduction
    2 Surface Albedo Modification Materials, Field Testing Deployment Methods, Instrumentation, Results, and Feasibility
    3 Modeling Study
    4 Modeling Results
    5 Conclusions and Discussions
    6 Future Work
    7 Modeling data
    Supporting Information

    Request permission
    Export citation
    Add to favorites
    Track citation

Share a link
Share on

    Linked In


The rising costs of climate change merit serious evaluation of potential climate restoration solutions. The highest rate of change in climate is observed in the Arctic where the summer ice is diminishing at an accelerated rate. The loss of Arctic sea ice increases radiative forcing and contributes to global warming. Restoring reflectivity of Arctic ice could be a powerful lever to help in the effort to limit global warming to 1.5°C. Polar ice restoration should be considered in planning of 1.5°C pathways. In this paper, a novel localized surface albedo modification technique is presented that shows promise as a method to increase multiyear ice using reflective floating materials, chosen so as to have low subsidiary environmental impact. Detailed climate modeling studying the climate impact of such a method reveals more than 1.5°C cooler temperatures over a large part of the Arctic when simulating global sea ice albedo modification. In a region north of Barents and Kara Seas temperatures have been reduced by 3°C and in North Canada by almost 1°C. Additionally, there are notable increases in sea ice thickness (20–50 cm Arctic wide) and ice concentration (>15–20% across large parts of central Arctic). These results suggest that the geoengineering technology proposed in this study may be a viable instrument for restoring Arctic ice.
Plain Language Summary

This paper describes a method to preserve and restore ice in the Arctic in order to reduce the effects of climate change. This method is benign by design, developed to restore ice in the Arctic in targeted areas to build back the reflective ice that has melted over the past several decades. The aim is to restore the Arctic ice's historic function of reflecting sunlight. By applying reflective materials such as glass microspheres on young, low‐reflectivity sea ice, we can protect the young ice from the summer Sun, much like a white shirt fends off the Sun for a person on a hot summer day. This way the ice may be conserved and converted over time into highly reflective multiyear sea ice. Climate modeling shows that this method can cool the Arctic significantly and can rebuild Arctic ice area and volume, hence reducing Arctic as well as global temperature rise.
1 Introduction

The rate of change in climate is much higher in the Arctic than elsewhere in the world, and as a result, the summer ice in the Arctic is diminishing at an accelerated rate (Intergovernmental Panel on Climate Change Fourth Assessment Report (IPCC AR4 (2007) et al., 2007; IPCC AR5, 2014)). Over the past few decades, roughly one half the surface area, and three fourths of the volume of ice in the Arctic, has now been lost, and what remains and is regrown each winter is now primarily young, and often first year, ice. Such a thin layer of ice transmits rather than reflects solar radiation, and melts quickly in the summer Sun. There is evidence as reported in a recent study that ice‐ocean albedo feedback in the Arctic is shifting to a seasonal ice zone (Kashiwase et al., 2017).

Given the above conditions, we are quickly transitioning to what is known as “Blue Arctic” with faster thinning of Arctic sea ice pack (Overland & Wang, 2013). An extensive climate modeling study based on the present‐day (business‐as‐usual) warming scenario suggests that there would be greater Arctic warming and sea ice loss using the latest climate model Community Earth System Model version 1 (CESM1; Community Atmospheric Model version 5, CAM5), with an ice‐free summer Arctic occurring by about 2060, as opposed to about 2100 in Community Climate System Model version 4 (CCSM4; Meehl et al., 2013). The loss of sea ice in the Arctic is accelerating positive feedback to global warming. In fact, Pistone et al., 2014 reports that over 30 years of available satellite measurements, the decrease in albedo averaged over the globe corresponds to a forcing that is 25% as large as that due to the change in CO2 during the same period. Wadhams (2017) reports an even larger value of about 50% that is added to the radiative forcing of the planet due to greenhouse emissions by changes in albedo alone, when taking into account both Arctic snow and sea ice retreat. Thus, restoring Arctic sea ice reflectivity appears to be a necessary and powerful lever that could help in limiting warming to 1.5°C. Excluding polar ice restoration could make the 1.5°C goal impossible to achieve, so considerations of polar ice should be included in 1.5°C pathways (IPCC Special Report on Global Warming of 1.5°C (IPCC SR15, 2018)).

A wide variety of climate mitigation and restoration strategies that broadly fall under the term geoengineering have been proposed in order to slow, stop, or mitigate the effects of global temperature rise. However, it is increasingly clear that efforts focusing specifically on the Arctic may be necessary in order to meet this goal (Stroeve et al., 2012). Several strategies have been proposed for preserving Arctic sea ice including winter ice thickening by wind‐powered pumps (Desch et al., 2017) and seawater brightening through hydrosols around the edges of the Arctic ice pack (Seitz, 2011). Cvijanovic et al. (2015) studied restoration of the Arctic ice via alteration of the ocean surface albedo. In that study, by keeping the ocean albedo high enough to maintain freezing temperatures they intensify the production of newly frozen ice and successfully expand the sea ice extent. The extending of the ice areal coverage is particularly important for reducing the polar amplification (accelerating the ice melt due to positive ice albedo feedback). However, their approach may not be very efficient in ice recovery since the generated young thin ice will still be highly transmissive of incoming solar radiation and will not insulate the underlying ocean as well as an otherwise thicker, brighter multiyear (MY) ice would have.

In this paper we report a low‐risk, localized, and potentially reversible restoration solution of rebuilding reflective MY Arctic ice through surface ice albedo modification. The albedo modification method has been developed with small, localized field testing of various candidate materials in various locations, including a test site in the Arctic. In our effort to understand and quantify the possible impact of such an intervention, we use climate modeling to model and simulate increasing ice albedo as an intervention to preserve and restore ice in the Arctic. The scientific question that we seek to answer in this paper is whether a surface albedo enhancement on Arctic sea ice impacts the Arctic sea ice pack, and if so what are the impacts of such an intervention on the climate in the short and long terms.

The paper is organized into six sections. In section 2 we present the details of this new restoration and the methods that are adopted in deploying and testing some albedo‐modifying materials. Section 3 details the climate modeling and simulation of this intervention, and in section 4 we focus on studying the results thereof. We summarize our results in section 5. Here we discuss the relevance of the results with respect to current research standards in climate modeling and geoengineering fields. We briefly discuss in section 6 the future work planned for studying specific areas to propose for targeted treatments that could be especially effective per unit area in preserving and restoring ice in the Arctic.
2 Surface Albedo Modification Materials, Field Testing Deployment Methods, Instrumentation, Results, and Feasibility

A localized surface albedo modification technique has been developed that shows promise as a method to slow Arctic melting, and as a means over time to increase MY ice using reflective floating materials, chosen so as to have low subsidiary environmental impact.
2.1 Materials: Overview

Dozens of materials have been considered and tested over the course of this research and development work, with a priority on safety, effectiveness, practicality, and reasonable cost.

The materials we have focused on most share many of the following characteristics:

    high albedo (preferably at least as bright as fresh snow);
    less dense than water (i.e., floatable);
    nontoxic (fed to fish and birds by an environmental testing lab with no ill effects);
    wettable (hydrophilic rather than hydrophobic, so the surface will not have an affinity for oil‐based pollutants);
    small diameter or thickness (allowing minimal materials usage per unit area);
    larger than respirable range (avoiding inhalation effects and nanoparticle concerns); and
    relatively low cost.

There are a number of commercially available materials that fulfill some or all of the above characteristics, including some kinds of sheet‐type materials, and some granular materials, including a class of commercial materials known as hollow glass microspheres. In layman's terms, these latter materials can be regarded to be chemically and physically like sand, but are floatable and round.

One of the granular materials on which Ice911 Research has done its small‐scale field testing in Minnesota and in the Arctic for ice and snow preservation is K1 microspheres, a member of the 3 M Glass Bubbles family of products. This product consists of bright white hollow glass spheres, of average diameter 65 μm, chosen for its brightness, nontoxic formulation, and relatively low cost. This material has a density ranging from 0.10 to 0.14 gm/cm3, with a typical value of 0.125 gm/cm3, and a particle size distribution (microns, by volume) of 10th percentile: 30 μm; 50th percentile: 65 μm; and 90th percentile: 115 μm.
2.1.1 Materials: Albedo Laboratory Testing

The albedo and light transmission properties of samples of the K1 glass microsphere material were tested using ultraviolet/visible/near‐infrared spectrophotometry by Covalent Metrology, an analytical laboratory. The samples were run on a Shimadzu Solid Spec 3700 DUV Spectrophotometer.

The K1 granular powder was tested at several different amounts/thicknesses, ranging from 102 to 331 μm in total, with the samples held between two glass slides. At an average granule diameter of 65 μm, these layer thicknesses cover a range from 1.5 to 5 monolayers of material, as shown in Figure 1. It can be seen in this figure that the transmission continues to be reduced and the albedo continues to increase as the thickness is increased, throughout the range tested.
Figure 1
Open in figure viewerPowerPoint
Laboratory testing results of (top) albedo and (bottom) light transmission for various layer thickness applications of K1 hollow glass microspheres.
2.1.2 Materials: Environmental Testing on Fish and Bird Species

The K1 material was tested by an environmental testing laboratory, EA Engineering, Science and Technology, Inc., PBC, on sheepshead minnows, Cyprinodon variegatus, with no ill effects.

EAG Laboratories tested the K1 material on northern bobwhites, Colinus virginanianus, also showing no deleterious effects.

See the supporting information Document, section S1, for further details on the environmental testing performed.
2.2 Field Testing Overview

To determine in the field how best to deploy and monitor various high‐albedo materials on ice and snow, field testing in various locations was done on a small scale, with all required permits and permissions, in locations in California including Serene Lakes in the Sierra Nevada range; Lake Miquelon, a Canadian lake near University of Alberta, Edmonton; a small man‐made pond in Lake Elmo, Minnesota; and North Meadow Lake, a shallow lake in the Barrow Experimental Observatory Area in Barrow, Alaska. The experience gained from the repeated seasons of experimentation in California, Minnesota, and Alaska led to improvements in the materials, instrumentation, and deployment techniques. It was found that wettable granular materials tended to be easier to maintain in place on the ice in windy outdoor conditions, especially if deployed in somewhat moist surface conditions, than sheet forms of materials. Deployment methods were developed for the granular materials for distribution by hand in very small areas, and for distribution with the aid of agricultural machinery for larger deployments on frozen lakes. The largest deployment area made by Ice911 to date was 45,000 ft2 (4,180 m2) of the 3 M K1 materials on a section of North Meadow Lake in Barrow, Alaska, in May 2017.

Albedo and ice preservation performance has been tracked using manual ice thickness measurements and automated cameras to record the performance of the materials and control areas, solar sensor pairs to determine albedo, strings of temperature sensors submerged in the ice and water below the instrumentation buoy, and tilt sensors. The buoys are powered by rechargeable batteries and solar panels for power. Various modes of wireless communication have been used, and data have been stored locally on SD cards in the instrumentation package on the buoy. An overview of the instrumentation challenges and strategies used in the Alaska testing was presented at a Linux conference in Europe (Chetty, 2017).

See the supporting information Document, section S2, for further details on the field testing overview.
2.2.1 Field Testing: Minnesota

Figure 2 shows the albedo measurements for a control area and an area treated with K1 material, during the time of the ice melt on a small pond in Lake Elmo, MN. Note that there are spikes in this real‐life real‐time data, possibly due to specular reflections from sunlight on the measured surfaces and that positioning the solar sensors can be challenging in the needed balance to ensure measurement of only the relevant area, while encompassing enough of the relevant area to be able to integrate over any heterogeneities in the surface. Throughout this time, the measurements show that the albedo values for the test area coated with the K1 are considerably higher than the untreated control area, adding 15–20% in increased reflectivity. Note that these albedo measurements are for the same areas and times shown in the automated photographs of Figure 4 taken during the ice melt.
Figure 2
Open in figure viewerPowerPoint
Albedo measurements for a control area and an area treated with K1 material, during the dates of the ice melt on the Minnesota pond in March 2016.

A literature reference on transmittance and albedo for thin young ice in the Arctic is shown in Figure 3 below (Taskjelle et al., 2016).
Figure 3
Open in figure viewerPowerPoint
Measured (a) albedo and (b) transmittance for the three days. The shaded area around each line represents the ±1 standard deviation for all measurements along the transect. The number of spectra was 42, 17, 11, and 14 for 17 April, 21 April, 25 April (bare ice), and 25 April (2 mm snow cover on ice), respectively. From Taskjelle et al. (2016)

Even with a thin coating of 1.5–5 monolayers of K1 applied, the 65‐μm diameter hollow glass sphere material has a higher albedo, from Figure 1, than does the 7 cm and upward of thin young ice albedo results shown in Figure 3.

Comparing Figure 3, showing albedo from field measurements of ice and snow mixtures (Taskjelle et al., 2016), with Figure 1, showing albedo from laboratory measurements of 1.5 to 5 monolayers of K1, we see that in laboratory testing, the 65‐μm diameter hollow glass sphere material has a higher albedo than does the ice and snow mix.

Comparing the albedo measurements from field measurements in MN (Figure 2) with field measurements of ice and snow mixtures (Figure 3), we see that the field‐tested areas treated with 5–10 monolayers of the K1 material show increased albedo as well.

Thus, in the laboratory and in the field, the treatments using hollow glass spheres contributed more reflectivity per unit thickness than the substantially thicker 7 cm ice layer of Figure 3.

To visually compare how the melt progressed in a treated and untreated areas, the photographic sequence in Figure 4 illustrates the melting in March of 2016 of one of the control areas and an area treated with K1 material on the test pond in Lake Elmo, MN. Melting and some refreezing over limited areas were underway by 5 March in parts of the treated and control areas.
Figure 4
Open in figure viewerPowerPoint
(a) K1‐treated area (left) and one of the adjacent control areas (right), on 5 March, 2016. (b) Same areas on 8 March, when there was a trace of ice left on one section of the control area. (c) 9 March, when all ice had melted from untreated control area (right). (d) 11 March and (e) later on 11 March, when the ice, which had persisted in the treated area for several days past the disappearance of ice in the untreated control area, finished its melt throughout the day.

By late 8 March (Figure 4b) there was a trace of ice left on one section of the control area.

With an almost identical experiment in the 2014–2015 season, on the day that the control segments completely thawed, about 3 inches (8 cm) of ice remained on the K1 treated segment, in spite of the center‐edge convective circulation. (No ice thickness data is available for the day that the control segments completely thawed in the 2015–2016 season.)

In the 2013–2014 ice season (the second year of experimentation on this pond), four not‐quite cubes wrapped with polyethylene film on the sides and the bottom were deployed before the pond surface froze, and after it was possible to walk on the ice, test materials were deployed. In the spring, it was observed that on the day the control section thawed completely, there was still 9″ (23 cm) of ice in the section treated with K1 glass bubbles (same application rate).

The demonstration of melting delay and of ice preservation is significant, especially when noting that the small test areas are heavily influenced by heat fluxes from the surrounding pond edges and the bottom of the pond. Consistent with this assumption were the manual measurements of a 3× increase in ice thickness observed in the 2013–2014 season (9″; 23 cm) versus the 2014–2015 season (3″; 8 cm) when the experimentation had included greater thermal isolation of each area, from the use of the plastic “cube” segmentation structure that reduced heat fluxes from other test areas, as well as the sides and the bottom of the pond.

Figure 5 illustrates that once all the ice has melted, the materials float on the water where they are blown to various obstacles, including the surrounding shore, and disappear.
Figure 5
Open in figure viewerPowerPoint
Same areas as were shown in Figure 4 (the previously K1‐coated area and the control), on 15 March 2016, after the materials have dispersed.

Speculations are that the hollow glass spheres may be assimilated into the mud at the sides of the pond, or eventually covered by other materials leading to sinking. Less probable but possible other routes to removal of the materials are dissolution of the silica hollow walls, or possibly even breakage. The eventual fate of the materials is still being investigated in the lab and being further studied in pond testing. Intentionally, broken samples have been studied previously for any evidence of sharp edges, and to date, none have been found. The investigation is ongoing. The overall result over time from any of these routes would lead to the minimal amounts of the sand‐like material that was applied (for instance, 5 to 10 monolayers of average 65‐μm diameter material) being added to the local environment in the mud at the sides or bottom of the pond.

The use of a high‐albedo material, such as the glass bubbles shown here, has been shown to modify albedo and to delay ice melt. This is one of several material possibilities that Ice911 has evaluated to date. Other ice albedo modification methods, such as a proposal for ice thickening (Desch et al., 2017), may provide alternative useful means to restore Arctic ice.

See the supporting information Document, section S3, for further details on field testing in Minnesota.
2.2.2 Field Testing Deployment Methods: Alaska

Ice911's Arctic field testing has been conducted, with appropriate field testing permits, over two seasons on sections of North Meadow Lake, a shallow lake in the Barrow Experimental Observatory Area in Barrow, Alaska. The greatest progress made in this testing in Alaska has been in ruggedizing the instrumentation and developing techniques for larger‐area (though still quite small‐scale) deployments of the albedo‐modifying test materials.

In the first season in Barrow, hollow glass spheres and similar granular materials were spread, as had been done on the pond in Minnesota, by distributing it by hand in premeasured amounts using a shaker over the test area. Localized wind shields were used to help the material settle in the desired location during application. Materials were distributed in roughly 5′ by 5′ (1.5 m × 1.5 m) areas, treated square by square to cover the total test area, as was done in Minnesota.

In the first deployment of the second season in Barrow, in March 2016–2017, the same distribution philosophy was still used, but with the addition of small portable greenhouses mounted on a pair of sleds to act as runners under opposite sides, to provide stronger and more effective localized wind shields during material application, in order to better weather the very strong winds at the site.

In May 2017, a second deployment was made, again on North Meadow Lake in Barrow, Alaska, over four areas labeled A, B, C, and D, again including K1 at two different loadings, another material, and a control area for comparison. This second deployment included the largest deployment to date by Ice911, covering an area of 45,000 ft2 (4,180 m2) of the 3 M K1 materials. This scaled‐up, but still small test, required a move beyond manual methods.

An agricultural seed spreader was used for the deployment, donated, and customized for Ice911's application by Truax Company, Inc. (Minnesota). The machinery allowed the materials to be delivered very close to the ground while the spreader was pulled along behind a snowmobile. The deployment area was shielded from the wind by flexible plastic sheeting attached to the spreader, segmented to add to the ability to follow the ground contours. The spreader was used to distribute various types of granular glass microspheres over several neighboring test areas on the pond. A thickness of 5–10 monolayers of material was used in the various areas of these field tests. The materials were successfully deployed, and after deployment the applications of granular materials seemed to be stable in the face of local wind and storms.

The ability of the materials to remain in place once applied, in the face of Arctic wind conditions, will be an important factor to monitor in further field work and observation. Because the materials are wettable, if they are applied when there is moisture present on the surface, they have a tendency to remain where they have been placed, held by the water and eventually frozen into the ice.

Various mechanical and communications problems precluded the quantitative data gathering on the 2017 field work in Barrow, Alaska. To address this, the Ice911 team has recently placed improved monitoring and communications equipment at the same location used in 2017, North Meadow Lake, to aid in rerunning the testing.

See the supporting information Document, section S4, for further details on field testing deployment methods used in Alaska.
2.3 Feasibility Study

Questions of practicality and feasibility are critically important in the development of the ice restoration solution detailed in this study. Our aim over time is to determine the most effective locations for placements of targeted and limited applications of materials in strategic locations. Such limited, targeted deployments would be desirable in terms of costs and logistics, rather than broader deployments, and are more likely to gain the required permits and permissions. It is thought that such applications in areas such as the Fram Strait to reduce ice export from the Arctic, and the Beaufort Gyre to rebuild the circulating MY ice, are likely to be particularly strategic areas for such treatments. An outline of costs and strategies for a deployment plan for a targeted treatment in the Beaufort Gyre is given in this section. Many more details can be found in the accompanying supporting information document.
2.3.1 Production and Cost of Microspheres

Current world demand for glass microspheres stands at approximately 68,000 t/year. A deployment of 25,000 km2 of microspheres, for example, at an average diameter of 65 μm and an applied layer thickness of 1 monolayer, or 65 μm, would require approximately 300,000 t of microspheres, which would consume world production by 2.2×. This materials demand necessitates a current microspheres producer to expand output, or for the deploying organization to build its own production facility. The production of 300,000 t of microspheres would cost approximately $300,000,000 at current prices ($12,000 per km2 of monolayer) and require a loading volume of approximately 400,000 m3.

Based on industry data, it would take approximately 200 production lines to produce 300,000 t of microspheres in one year. The raw material required, silica, is in ample supply to produce this volume of material and is thus not a limiting factor to production at this quantity. Microsphere production facilities currently produce with demand and affirm that production can be rapidly increased to meet 300,000 t of demand given the capital for expansion.

The lowest‐cost location to produce microspheres is currently in Shandong, China, where one of Ice911's microsphere suppliers can currently produce up to 45,000 t/year and has access to the seventh busiest port in the world, Qindao Port. Qindao Port is located on the Yellow Sea and has the infrastructural capacity to load a very large crude carrier (VLCC) ships with microspheres. A factory sited near railroad tracks could deliver material in 18‐t rail cars directly to the port for loading on a VLCC.
2.3.2 Shipping of Microspheres to Beaufort Gyre

Once the material is loaded onto the 1–3 VLCC ships at Qindao Port, the ships follow an approximately 8,000 km trajectory to get to Prudhoe Bay/Barrow, Alaska, and the Beaufort Gyre. This trip takes approximately nine days at an average cargo carrier speed of 20 knots. Shown below is a map of the ship's ~8,000 km path. The VLCC ships can travel approximately 46,300 km without refueling, so they are able to go directly from the port in China to the Beaufort Gyre, deploy a full load of material, and go back to the port to refill. With one blower chute on the ship, deployment of the full load would take an estimated one day—meaning that the process from port departure to full deployment is less than two weeks. The relatively short deployment process allows the option to reduce capital costs by only purchasing one VLCC and having it deploy the material in three separate trips. However, three VLCCs deploying at the same time would assure the most uniform distribution of material and would not be subject to potentially large time differences between trips due to inclement weather.

The approximate cost of shipping the material from the factory to Qindao port at current Chinese rail freight prices is around $10.4 million dollars. The approximate fuel cost to send one VLCC on the 8,000 km path is $440,000 at current fuel prices.
2.3.3 Distribution of Microspheres

To achieve a nonlinear impact on global climate (with a particularly effective deployment area), we explore the idea that the glass microspheres be distributed over a strategic area of sea ice approximately 15,000 to 100,000 km2 in the Fram Strait, to acts as a flow restrictor on Arctic ice and water currents, and/or the Beaufort Gyre, which acts as a flywheel for Arctic ice.

The optimal time of season for deployment is estimated to be in October and November during the grease ice phase of ice formation. Grease ice is the beginning phase of ice formation in which the ocean surface layer has the characteristics of a slush. New ice forms as ice frazil crystals grow in a layer of grease ice. The grease ice layer is measured to be as thick as 70 cm but is typically around 10 cm and eventually congeals into a layer of solid ice, depending on surface temperatures (Smedsrud & Skogseth, 2006). Application of the microspheres to grease ice ensures that microspheres will be applied to the top layer of what will become first‐year (FY) ice, which cools the surface layer and enhances the reflectivity of FY ice as it builds from below. Moreover, application to grease ice circumvents any interaction of the material with mammals such as polar bears as it is dispersed from the deployment ship because they are unable to access grease ice.
2.3.4 Dispersal Vehicle for Material Deployment

In order to deploy microspheres in the harsh Arctic environment, various methods of material distribution in the key locations have been explored:

    Land‐based methods such as custom terrestrial spreaders have been used in Ice911's test site in Barrow, Alaska, but will not be sufficient to cover these large areas in reasonable time. The terrain on sea ice is often very nonuniform and can change rapidly, making such an operation hazardous to people and equipment as well.
    Air‐based methods, such as adapting drones, balloons, or aircraft, quickly run into limitations of loading, particularly from the large volume needed of the very lightweight materials—thus requiring considerable fuel and many flights to accomplish the job.
    This document details the use of one or more large vessels on the ocean itself, with the range of material distribution extended by the use of blowers.

The optimal vehicle for deployment must combine long travel range, durable construction, and immense cargo volume due to the packing density of the material. Therefore, Ice911 has determined that the most feasible route to full‐scale deployment is to use readily available ships. Optimizing for lowest cost, the deploying organization can buy one to three VLCC ships and adapt them for the deployment of microspheres. VLCCs are ocean tankers built to carry large volumes of cargo or liquid internationally. With cargo volumes between 150,000 and 200,000 m3, it would take either three VLCCs or one VLCC on three separate trips to deploy 25,000 km2 of microspheres.

See the supporting information Document, section S5, for further information strategies for materials dispersal for targeted deployment in the Beaufort Gyre.

In the next two sections, we describe the details of climate modeling of the Ice911 ice preservation and restoration technology and further study the impacts of the albedo modification material on the climate.
3 Modeling Study

A detailed modeling study is designed to simulate the potential impact of Ice911's albedo modification through added material on the Arctic sea ice, using a state‐of‐the‐art climate model. It is aimed at investigating regional‐ and large‐scale climate impact of changes in the Arctic sea ice albedo. We specifically seek to understand how these changes impact the distribution of the Arctic sea ice area, concentration, volume, thickness, and temperature.

We assume that the major physical effect of the Ice911 material when applied to sea ice is to increase the sea ice albedo. We investigate the effect of increasing of the Arctic sea ice albedo basin‐wide by changing the model sea ice albedo everywhere where sea ice exists (both Arctic and Antarctic). This simulation will provide the extreme scenario and show the potential impact when the entire sea ice pack is treated with Ice911 material.

Our major tool is the National Center of Atmospheric Research CESM (version 1.2). This is a fully coupled global climate model, which consists of atmospheric, ocean, land, and sea ice components interacting and exchanging water and energy fluxes via coupler (CPL7; Hurrell et al., 2013). We utilize B compset, which is a present‐day (2000s) greenhouse emission forcing scenario. It is configured with components CAM4, CLM4 (Community Land Model, version 4), POP2 (Parallel Ocean Program, version2), and CICE4 (Los Alamos Sea Ice Model, version 4) onto a nominal grid resolution of 0.9 × 1.25° in the atmospheric and land model domains and 1° in the sea ice and ocean model domains.

To modify the sea ice albedo, we are using the “delta Eddington” (DE) shortwave parameterization in the sea ice model component of CESM (Briegleb & Light, 2007), which calculates the albedos and the absorption of the shortwave fluxes. Figure 6a shows the modeled ice albedos by the DE (black lines) parameterization compared to SHEBA observations of the summer of 1998 (red and blue lines). The relationship between the albedo and the ice thickness seen in Figure 6b for both observations (squares) and models (CESM‐DE in red line, CCSM3 in blue line) is strongly nonlinear up to 1‐m‐thick ice. For 100 μm of ice, albedo is 0.07, for 300 μm about 0.008 from various researchers, and the highest albedo of about 0.65 is for 1 m of ice. The laboratory‐tested value of K1 of approximately 0.45 for 330 μm of K1 as was shown previously in Figure 1 is far higher than the values of 0.008–0.07 found by various researchers for 100–300 μm of ice. In fact, the albedo of 0.45 measured or 330 μm of K1 material is closer to the values measured for over 0.2 m of ice, 600 times the thickness of the albedo enhancing material used.
Figure 6
Open in figure viewerPowerPoint
delta Eddington (DE) parameterized ice albedo: (a) for different wavelengths compared to observations and (b) for different ice thickness compared to observations (squares) and albedo parameterization in earlier model version CCSM3. From Briegleb and Light (2007). Copyright 2007 Bruce Briegleb and Bonnie Light. This work is licensed under a Creative Commons Attribution‐Noncommercial 4.0 International License (

There are three types of physical media considered in the sea ice model: sea ice (represented with five ice thickness categories), snow, and melt ponds. For each of them there is a separate calculation for albedo and shortwave absorption involving their optical and thermal properties, which can be changed via tunable parameters. In a series of test experiments and also previous studies (Urrego‐Blanco et al., 2016) we found that the highest sensitivity of the broadband albedo is to the snow tuning parameters. By tuning the snow parameters we create the perturbations in overall broadband albedo, which we use to set up our final sensitivity experiment.
3.1 Numerical Experiments

As a starting point, a common spin‐up data set is used for all simulations. This is necessary to obtain a reasonably steady stable climate state, which will provide the initial conditions for all the simulations that we need to run. After 40 years of integration the atmospheric state seems to tend to equilibrate. All of the biases in the radiative balance are reduced. The ocean takes longer to spin‐up and is still adjusting at 40 years. However, as a first approximation, we use this model solution to initialize our next set of experiments.

We carried out two major experiments: (1) Control Experiment where we keep the standard settings of the model. It starts in 2000 and represents a short‐term (ending in 2040) future projection of the current climate state under the 2000s greenhouse gas (GHG) forcing. (2) Global Sea Ice Albedo Experiment where we change the sea ice albedo everywhere sea ice exists (both Arctic and Antarctic). The albedo perturbation is introduced from the beginning in 2000 and is applied continuously during the entire integration. This sensitivity experiment is representative for the application of the geoengineering ice restoration technology of Ice911 on the global sea ice pack.

The albedo perturbations, which we set up in our experimental design, are presented in Figure 7. In Figures 7a and 7b we see the mean state of the surface albedo in the Arctic area for the 40‐year period of integration in the two model simulations considered in this study (Global and Control, respectively). Both spatial patterns look similar (Figures 7a and 7b) with the highest albedo values in the area of land‐ice covered Greenland and over the thickest sea ice pack in central Arctic close to the Canadian Archipelago. The lowest surface albedos are found over the ocean and land, which have less reflectivity than the ice surfaces. Within the Arctic basin, the lowest albedos of 0.2–0.25 are found in the marginal ice zone rapidly increasing toward the Central Arctic up to 0.7–0.75. The albedo perturbations in the sensitivity experiment are applied only on the sea ice cover so we see the largest albedo differences between the two experiments over the Arctic sea ice pack (Figure 7c). The Global ice albedo ranges from 0.25 to 0.8 within the Arctic basin. The strongest impact of the albedo modification is found in the marginal sea ice zone.
Figure 7
Open in figure viewerPowerPoint
Arctic surface albedo perturbations: (a) Map of Arctic surface albedo for Global albedo modification of sea ice averaged over 40 simulation years. (b) Map of Arctic surface albedo for Control averaged over 40 simulation years. (c) Map of differences in surface albedo between Global albedo modification and untreated Control averaged over 40 simulation years. (d) Time series and linear trends of Arctic annual sea ice albedo for Global albedo modification (red), and untreated Control (blue).

It may be noted that the range of values for Arctic surface albedo in our Control experiment (Figure 7b) is within the same range as the ones reported by Briegleb and Light (2007) (Figure 6). Figure 7d shows the annual time series of the average Arctic sea ice albedo and their linear trends for the two experiments. In the Control experiment (blue) we see a rapid decline of the surface albedo within the next 25 years. The latter is a consequence of the ongoing transition to “Blue Arctic” state of thinner, faster, and less compact Arctic sea ice pack (Overland & Wang, 2013). In the sensitivity experiment, Global, where the albedo modification is applied, we see a modest long‐term positive trend in the sea ice albedo. The magnitude of the Arctic averaged sea ice albedo increased 10–15% compared to the Control. This modeled albedo modification is comparable to the Ice911 K1 material albedo enhancement of 15–20% in the Minnesota pond field experiment (Figure 2).
4 Modeling Results

In the following section we present the impact of the surface sea ice albedo perturbations, which we set up in the Global sensitivity experiment to simulate the effect of the Ice911 geoengineering on the Arctic sea ice area, volume, and surface temperature. We use the results from the Control experiment, projection of the current climate state under GHG 2000s forcing, to provide a baseline for comparison. Note that the only differences between the two experiments are the albedo perturbation.
4.1 Sea Ice Area and Concentration

We begin our analysis with the impacts of the albedo perturbation on the seasonal variability of the Arctic sea ice area. The total sea ice area is derived as cumulative sum of the ice‐covered area of all grid cells of the model grid within the 15% contour of sea ice concentration in the northern hemisphere. It is expected that the largest effect of the enhanced albedo will be on the thinner, such as younger or melting, sea ice, due to its lower reflectivity compared to the thicker MY sea ice. The annual cycles of the total Arctic sea ice area in the two numerical experiments shown in Figure 8 reveal that the most significant increase of the ice area is during the melt season.
Figure 8
Open in figure viewerPowerPoint
Annual cycle of the Arctic total sea ice area for Global albedo modification (red) and untreated Control (blue).

The seasonal maps of the sea ice concentration (Figure 9) confirm this result showing modest increase during the winter at the sea ice edge most of all in the Barents Sea (Figures 9a and 9b) and strong increase of more than 20% of the ice concentration everywhere in the Arctic marginal ice zone during the summer (Figures 9c and 9d).
Figure 9
Open in figure viewerPowerPoint
Sea ice concentration maps of (a) winter mean (January‐February‐March) for Global albedo modification; (b) difference in winter (January‐February‐March) mean between Global albedo modification and untreated Control; (c) summer mean (July‐August‐September) for Global albedo modification; (d) difference in summer mean (July‐August‐September) between Global albedo modification and untreated Control.

The interannual variability and long‐term trends of the Arctic total sea ice area are shown in Figure 10. The albedo perturbation has caused an instantaneous increase of the sea ice area during the initial period of the integration of Global experiment from 9 × 1012 to 9.7 × 1012 m2. As anticipated for a present‐day scenario, the Control case has a long‐term decline of the annual total ice area. In order to compare our results to the reported in the IPCC AR5 we consider the sea ice extent, which is the overall surface area within the 15% contour of sea ice concentration. It is often preferred to the actually ice‐covered area in previous studies (including IPCC AR5) due to the higher confidence in measurement and simulation. For the purposes of the current numerical study we focus on the sea ice area since that it is the actual area where the sea ice albedo perturbations are applied. In our Control experiment the September minimum of the total Arctic sea ice extent (not shown), often used as an indicator of transition to seasonal ice‐free Arctic, has a decreasing trend of ~ −0.2 × 106 km2/decade, which is smaller than the observed −0.8 × 106 km2/decade for the same period but within the range of the CMIP5 historical simulations (IPCC AR5, Chapter 9, Figure 9.24). In the albedo perturbation Global case we see increasing trends in both annual and September ice time series.
Figure 10
Open in figure viewerPowerPoint
Time series and linear trends of Arctic annual total sea ice area for Global (red) and Control (blue).

For a stable Arctic recovery it is important the FY ice to survive the melt season and transition to a MY sea ice. To see if that is the case in our albedo perturbation experiment, we investigate the ratio of MY and FY ice fraction. In Figure 11, we see that in the Control case, this ratio is decreasing, meaning that the fraction of the MY ice is reducing compared to the FY fraction. This type of FY dominant ratio suggests transitioning toward seasonal ice pack when the generated winter sea ice does not survive the following melt season. In contrast, the Global experiment shows a strong increasing trend of the MY/FY ratio proving the growth of the perennial ice fraction.
Figure 11
Open in figure viewerPowerPoint
Time series and linear trends of the ratio between multiyear and first‐year sea ice fraction in Global (red) and Control (blue).
4.2 Sea Ice Volume and Thickness Changes

The sea ice volume is a product of the sea ice area and sea ice thickness; thus, its changes reflect both the changes in the sea ice thickness and the sea ice area. To evaluate the fidelity of our Control case Arctic sea ice volume, we compare it to the Pan‐Arctic Ice Ocean Modeling and Assimilation System (PIOMAS) Arctic Ice Volume Reanalysis for the period 2000–2017 (Zhang & Rothrock, 2003). PIOMAS was developed at Applied Physics Laboratory with the goals to simulate and understand the mechanisms of the current Arctic decline and project the future Arctic states under different scenarios and so far it is the only sea ice data assimilation product and closest to observational estimate available for the sea ice volume. Figure 12 shows the monthly time series of the total sea ice volume in Control and PIOMAS. Their magnitude of the seasonal cycle seems to be in a good agreement, and they, both exhibit declining trend in this period, but it is much smaller in the Control. There is a better agreement for the later period 2010–2017 where comparison of the difference in the average sea ice volume in our Control for the year 2017 is about 1,000 km3 higher than in 2010, which is in the same order of magnitude as has been reported (1,100 km3) in the following PIOMAS model literature‐sea‐ice‐volume‐anomaly/.
Figure 12
Open in figure viewerPowerPoint
Time series of total Arctic sea ice volume of Control (blue) compared to PIOMAS data (red).

We compare next the annual cycles of Arctic total sea ice volume between the two numerical experiments Control and Global shown in Figure 13. Clearly, the Global albedo modification experiment shows larger total Arctic sea ice volume compared to the Control. Compared to the changes in the sea ice area where the largest impact from the albedo modification was in the summer season, the ice volume is enhanced through the entire year.
Figure 13
Open in figure viewerPowerPoint
Annual cycle of the total Arctic sea ice volume in Global albedo modification (red) and untreated Control (blue).

The maps in Figure 14a shows the winter mean state of the Arctic ice thickness distribution during the 40 years of integration in the two experiments. They both simulate the observed pattern of thicker ice in the Canadian Archipelago coast and Greenland, thinning toward the Siberian coast, and the marginal zones in the North Atlantic and Pacific varying in 0–5 m range. The differences between the two experiments (Figure 14b) demonstrate large‐scale thickening of the ice pack in the case of increased albedo (Global exp) ranging from 20 cm up to 1 m in the Canadian Archipelago. The impact is even stronger in the summer season where the central Arctic has thickened with about 20–50 cm and gotten covered with >1.5‐m‐thick ice (Figures 14c and 14d).
Figure 14
Open in figure viewerPowerPoint
Sea ice thickness maps of (a) winter (January‐February‐March) mean for Global albedo modification; (b) difference in winter mean between Global albedo modification and untreated Control; (c) summer (July‐August‐September) mean for Global albedo modification; (d) difference in summer mean (July‐August‐September) between Global albedo modification and untreated Control.

The time evolution of the Arctic averaged annual sea ice volume is shown in Figure 15. In the Control (blue) we see a declining trend, which is similar to that seen in the PIOMAS sea ice volume (see Figure 12) over the period 2010–2017 (see‐sea‐ice‐volume‐anomaly/).The continuous application of the Ice911 material causes ice volume increase of ~0.5 to 1%/year. For the 40‐year period, the Global experiment shows 6% increase per decade while the Control has 2% decrease; therefore, the albedo enhancement experiment mitigates the Arctic sea ice volume by 8% per decade.
Figure 15
Open in figure viewerPowerPoint
Time series and linear trends of the Arctic annual total sea ice volume for Global albedo modification (red) and untreated Control (blue).
4.3 Surface Temperature Changes

Changing the reflective properties of the sea ice will lead to changes in the Arctic energy balance, affecting the surface energy budget thus changing surface temperatures. The detailed consideration of the energy budget results is out of the scope of the current paper. Here we consider the impact of the albedo increase on the regional Arctic and world‐wide global surface temperature (Figure 16). In Figure 16a, we show the maps of the surface temperature mean climatologies for the 40‐year period of integration. The cooler temperatures are seen over the most reflective ice areas. The difference map in Figure 16b reveals that increasing the albedo in Global experiment is causing a decrease in the surface temperature. Over a large part of the Arctic we see more than 1.5°C cooler temperatures. In a region north of the Barents and Kara Seas temperatures have decreased by 3°C and in North Canada by almost 1°C.
Figure 16
Open in figure viewerPowerPoint
Surface temperature maps of (a) annual mean for Global albedo modification (b) difference in annual mean surface temperature between Global albedo modification and untreated Control.

Figure 17a shows a long‐term declining trend of the Arctic surface temperature in the Global experiment. Such an Arctic basin‐wide cooling trend in temperature over the long run may prove to be promising for Ice911's geoengineering technology, if it can indeed be tested and demonstrated: (i) at large scale, (ii) for extensive duration, and (iii) in maritime environments. So far the results are presented here only for an idealized scenario involving albedo modifications to the entire global sea ice area. Figure 17b shows a long‐term warming trend of about 1°C of globally averaged surface temperature interannual anomalies for the untreated Control (blue dashed line) and a modest cooling trend of about 0.02°C for the Global albedo modification case (red dashed line), over the period 2000–2040. The Control case uses 2000 GHG forcing for each year of the simulation. We assumed that the emission level at year 2000 is not increasing with time (as if due to stricter emission enforcement already in place). The results from this scenario are similar to Representative Concentration Pathway (RCP) 2.6. This shows that over the period 2000–2040, Global albedo modification could in fact prevent the 1°C rise in average global temperature predicted by the untreated Control, and in addition could further reduce the average global temperature by 0.02°C during the same period.
Figure 17
Open in figure viewerPowerPoint
(a) Time series of Arctic annual mean surface temperature and (b) global averaged surface temperature interannual anomalies (with climatological mean removed) for Global albedo modification (red) and untreated Control (blue).
5 Conclusions and Discussions

The novel idea that this paper presents is to explore whether melting of sea ice in the Arctic can be slowed down by using a localized reversible geoengineering Arctic restoration technology that increases Arctic sea ice albedo. It addresses the scientific question of what is the impact of albedo enhancement over sea ice particularly over the Arctic. It shows how such a modification impacts the sea ice area, sea ice volume, and temperature and demonstrates the nature of these responses.

Seasonal sea ice area shows significant increase compared to the untreated Control during the melt season. The seasonal sea ice concentration shows a modest increase during the winter at the sea ice edge and a stronger increase in a larger marginal ice zone during the summer. Increasing the sea ice albedo results in increased ice concentration or more compact ice pack, particularly during the summer melt season in the marginal ice zone.

The continuous application of the Ice911 material causes ice volume increase of (~0.5 to 1%/year). Sea ice volume increases are manifested in large‐scale thickening of the ice pack in the case of the increased albedo experiment, ranging from 20 cm up to 1 m in increased ice thickness in the Canadian Archipelago.

Climate modeling shows that a large part of Arctic sees more than 1.5°C cooler temperatures for global albedo modification as compared to the untreated control case. In a region north of the Barents and Kara Seas temperatures are reduced by 3°C and in North Canada by almost 1°C. From an analysis of the global average surface temperature anomalies trends, it is seen that over the period 2000–2040, the Control shows a 1°C rise in temperature, whereas the Global albedo modification case shows a modest decrease of 0.02°C over the same period. This demonstrates that global albedo modification could not only prevent the otherwise‐expected 1°C rise in average global temperature but could further reduce the average global temperature by 0.02°C over the same period. These results are a promising indicator of the mitigation potential that exists from prospective technologies that are capable of enhancing surface albedo of sea ice over vast geographic areas.

An albedo enhancement in the Arctic is a relatively new idea that needs to be explored with further research. Let us discuss some of the pertinent issues about the set of experiments reported here and compare our results to standard benchmarks. Our simulation is initialized using GHG forcing from year 2000 using CESM compset “B_2000.” Our study is in essence a perturbation experiment. We are comparing a relatively realistic Control (using the greenhouse forcing from year 2000) to our test case of enhanced Albedo (which is identical with the untreated control in every way except for the enhanced albedo). We have compared our Control to the PIOMAS (Schweiger et al., 2011, Zhang & Rothrock, 2003) data for the period 2000–2017 (see Figure 15). We note that the average sea ice volume time series in the control run shows a decreasing trend with time, in general, which is similar to that seen in the PIOMAS sea ice volume over the period 1980–2017.

It is worthwhile to briefly note here some of the other climate intervention technologies that are currently being explored as possible options. Bright hydrosols (Seitz, 2011), marine cloud brightening (Latham et al., 2012, 2014), blowing water over the Arctic sea ice (Desch et al., 2017), stratospheric aerosol injections, and experiments in GEOMIP (Kravitz et al., 2015; Tilmes et al., 2016) are some of the climate intervention technologies being reported in the recent literature. Seneviratne et al., 2018 reports a comparison of such technologies that may be grouped either as global solar radiation management (SRMglob) methods and regional land radiative management (LRMreg) methods with details of benefit and limitations for each method. The regional land radiative management described in Seneviratne et al. (2018) involves increasing the land surface albedo over HAA (Human‐affected and affecting) regions. The global‐scale temperature response from their experiment for such an application over the whole HAA regions is about 0.7°C particularly in the context of a 3.7°C warming for the 4 × CO2 scenario, but negligible when only applied over single regions (see their Figure 2).

In order to quantify the changes in the temperature, ice area, and volume brought about by the albedo modification technique described here, we attempt to compare in particular to results from the CMIP5 set of emission‐specific scenarios like the RCPs (see discussion in previous section). We note that our results are not directly comparable to the CMIP5 given the differences in the scenarios and metrics. However, a comparison from the IPCC AR5 report suggests that our Control simulation follows trends similar to RCP2.6; for example, our Control shows a 1°C increase in average global surface temperature between 2000 and 2040 (see IPCC AR5 report, A similar comparison holds for the September sea ice extent between our Control and the RCP2.6 scenario reported in IPCC AR5. The global albedo enhancement in our Global experiment demonstrates a temperature mitigation potential of 1.02°C and sea ice volume mitigation of about 8% per decade during the period 2000 to 2040.

Some specific advantages of the Ice911 approach, and ice rebuilding efforts in general, are that it rebuilds a natural state of ice in the Arctic. If the treatment is discontinued, there is no predicted rebound effect. The technology can be developed and studied with the aid of small, permitted field testing. The materials used in the Ice911 treatment are considered nontoxic, consisting of sand component silica, and in testing to date, they have shown no adverse impact on wildlife. The risks of an ice‐free Arctic include the Arctic's contribution to increased world‐wide temperatures through absorption rather than reflection of incoming summer‐time solar radiation This in turn could lead to methane release from frozen permafrost reserves that are no longer protected by overlying Arctic sea ice. These serious climate risks are predicted to be greatly reduced by the restoration of thicker Arctic ice.

These advantages are in contrast to some apparent challenges that have been identified in some other proposed geoengineering techniques, such as the potential for damage to the ozone layer from deploying stratospheric sulfate aerosols for solar radiation management (see Seneviratne et al., 2018).

Simulating concurrent and continuous application of the Ice911 albedo enhancing material shows an immediate shift of the Arctic climate state to a thicker and colder Arctic. Using intelligent geoengineering (a repeating cycle of model, test, and monitor), with safe materials and a localized deployment, preliminary results indicate that a basin‐wide albedo modification in the Arctic has potential to reduce climate change effects by thickening ice and lowering temperatures in the Arctic.
6 Future Work

Further study to establish the safety and risks of applying any material to restoring Arctic ice is needed. This includes further characterization of the ultimate fate of such added materials in polar environments—what proportion of the material sinks, floats, or dissolves over time is an area of active investigation by our team. While testing of the materials on fish and birds has shown no ill effects, evaluation of potential impacts by the materials on some further key species, such as marine mammals, needs to be done.

We also plan further work on deepening and quantifying our understanding of the albedo and thermal effects of the deployed materials on ice and snow. This could include addition of some periodic measurements using albedo sensors mounted on drones. We plan to incorporate redundant sensing architectures that prevent a single point of failure, which could, for instance, guard against the potential effects of a battery malfunction in a remote buoy. We also plan to include robust multisite data acquisition to gather more high‐quality quantitative data on ice thickness, in addition to the data from image sensors, temperature strings, and manual readings of ice cores.

In addition, we want to conduct a thorough study through climate modeling to see if the process of increasing sea ice albedo in the Arctic and consequently increasing the sea ice pack and cooling down the temperature in Arctic could lead to any adverse climate effects anywhere in the world, particularly examining the impact of storm tracks and extreme weather events. As weather events have become more severe with the disappearance of much of the historic levels of multiyear ice in the Arctic, we need to explore through further climate modeling whether restoring ice in the Arctic could lead to a reduction of the severity of such extreme weather events.

Further work is planned for studying specific areas to propose for targeted treatments, areas that could be especially effective per unit area in preserving and restoring ice in the Arctic. Toward that end, our next step will be to model and simulate the targeted deployments of Ice911 materials over strategically chosen smaller areas in the Arctic and study their climate impacts. We seek areas that have a particularly large impact on ice restoration and cooling. In actual implementation of the Ice911 ice restoration geoengineering solution, starting with such areas would be of crucial importance from the standpoints of effectiveness, logistics, and cost minimization.
7 Modeling data

The link to the publicly accessible data that Climformatics used to generate the climate modeling plots in this paper is

Ice911 is a registered 501(c)3 nonprofit, and wishes to express sincere thanks to all its individual, foundation, and corporate donors and supporters, its many volunteers, board members, advisory board members, and other key advisors. Special thanks go to the LaureL STEM Fund, The Mendelsohn Family Foundation, Barbara Grewen, and many other organizations and individuals for vital support. Without the contributions of these many individuals and organizations, this work could not have been done. We thank several corporations and nonprofits for key donations in kind and opportunities: Alaska Airlines, 3 M, Amazon Web Services, Covalent Metrology, Domino Data Lab, Google for Nonprofits, GSV Labs, Sustainable Silicon Valley, and the Truax Company, Inc. for a customized agricultural spreader. We thank the Cleantech Open, in which both Ice911 Research and Climformatics participated, resulting in many benefits, including meeting opportunities that led to the current collaboration. We thank IBM and Sabalcore for their generous donations of supercomputer time and administrative support and CESM‐NCAR for use of their climate modeling tools. We thank Satish Chetty for expert instrumentation and deployment advice, Don Perovich and Phil Rasch for expert ice albedo and modeling advice, Peter Wadhams for his valuable insight into albedo observations, and Jim Haywood for kindly providing us valuable insights.

Supporting Information

The capital expense figures in the Supporting Information assume a 30 micron material diameter; this information was omitted from the originally published version of the Supporting Information. The following sentence has been added to the Supporting Information: “A commercially available version of the material with a 30‐micron diameter and higher density is assumed for this targeted deployment calculation, to optimize required number of ships, logistics and cost.” This may be considered the authoritative version of record.


    Follow journal


Back to Top
AGU Journals Logo

    AGU Publications
    AGU Membership

© 2019 American Geophysical Union
Additional links
About Wiley Online Library

    Privacy Policy
    Terms of Use

Help & Support

    Contact Us


    Subscription Agents
    Advertisers & Corporate Partners

Connect with Wiley

    The Wiley Network
    Wiley Press Room

Copyright © 1999-2019 John Wiley & Sons, Inc. All rights reserved
Wiley Home Page
Pages: [1]