COOPER ISLAND IS a small low-lying barrier island a short distance off Point Barrow, Alaska, the northernmost point of the United States and indeed of North America. It sits some three hundred miles north of the Arctic Circle and is the nesting and breeding site of a colony of black guillemots, a not-so-common seabird of the Arctic. In 1972, a young ornithologist named George Divoky began a study of the breeding habits of these birds.43 Every summer, for the next thirty years, he spent on Cooper Island with the black guillemots and an occasional polar bear. Most summers were in "solitary confinement," but occasionally he took a field a.s.sistant. Carefully he noted the dates when the guillemots returned to Cooper, the dates when they laid their eggs, the dates when chicks hatched and later fledged. What he has discovered is that the entire reproductive sequence has shifted more than ten days earlier in the Arctic summer. Guillemots will nest as soon as the snow melts, but no sooner, and so the earlier nesting of these seabirds serves as a proxy for the timing of the annual snowmelt. But there was also some bad news for the guillemots: their population began to diminish around 1990. Year by year the Arctic warming has been moving the sea ice much farther from the nests on Cooper Island. Because the margins of the sea ice are favorite feeding spots for these seabirds, the retreat of the sea ice has been slowly moving food almost out of their reach. Every summer, for the next thirty years, he spent on Cooper Island with the black guillemots and an occasional polar bear. Most summers were in "solitary confinement," but occasionally he took a field a.s.sistant. Carefully he noted the dates when the guillemots returned to Cooper, the dates when they laid their eggs, the dates when chicks hatched and later fledged. What he has discovered is that the entire reproductive sequence has shifted more than ten days earlier in the Arctic summer. Guillemots will nest as soon as the snow melts, but no sooner, and so the earlier nesting of these seabirds serves as a proxy for the timing of the annual snowmelt. But there was also some bad news for the guillemots: their population began to diminish around 1990. Year by year the Arctic warming has been moving the sea ice much farther from the nests on Cooper Island. Because the margins of the sea ice are favorite feeding spots for these seabirds, the retreat of the sea ice has been slowly moving food almost out of their reach.

NATURE'S BEST THERMOMETER, perhaps its most sensitive and unambiguous indicator of climate change, is ice. When ice gets sufficiently warm, it melts. Ice asks no questions, presents no arguments, reads no newspapers, listens to no debates. It is not burdened by ideology and carries no political baggage as it crosses the threshold from solid to liquid. It just melts.

THE ICE SEASON.

When I was a boy growing up in eastern Nebraska, the calendar of certain activities was set by the seasons. Neighborhood hockey started up when nearby George's Lake froze over, and duck hunting began as the myriad channels of the Platte River became choked with ice. The family springtime fishing trip to the boundary waters of Minnesota was determined by the breakup of the winter ice six hundred miles to our north. In late May my father would be on the phone to friends in International Falls, Minnesota, checking whether the ice had moved out, and one week after the breakup, we were there trolling for walleyes. The rhythms of communities the world over have similarly been tied to the comings and goings of the annual ice.

Madison is the Wisconsin state capital and home to the University of Wisconsin. The city sits between Lakes Mendota and Monona, bodies of water that have provided recreational activities for residents ever since the city was founded in 1836, the same year the Wisconsin Territory was created. Perhaps not surprisingly, the dates of fall freezing and spring breakup have been dutifully recorded for almost a century and a half, and they tell a very interesting story. In 1850, Lake Mendota froze in early December and broke up in early April, but 150 years later, freezing had shifted to some nine days later and breakup occurred almost two weeks earlier.

Along the eastern sh.o.r.e of Lake Michigan is Grand Traverse Bay, another location with diligent record-keepers since the mid-nineteenth century. The long record compiled there shows that since 1851 the bay froze over completely at least seven times in each each of the first twelve decades; this figure dropped to six times in the 1980s, three times in the 1990s, and only twice in the first decade of the twenty-first century. of the first twelve decades; this figure dropped to six times in the 1980s, three times in the 1990s, and only twice in the first decade of the twenty-first century.44 For the years when the bay has frozen over, the period of winter ice cover has diminished by thirty-five days. For the years when the bay has frozen over, the period of winter ice cover has diminished by thirty-five days.

But it is not just lakes in North America that are showing trends toward shorter intervals of annual ice cover-in Scandinavia and Europe, in Asia and j.a.pan, the long-term observations are telling the same story. And it is happening not just to lakes-major rivers leading to the Arctic Ocean, such as the Mackenzie in Canada and the Angara and Lena in Siberia, show similar trends. Wintertime ice in the freshwater of the Northern Hemisphere is becoming a much rarer commodity.45 Mountain glaciers everywhere-in New Zealand, the Andes, the Alps, Alaska, the Rocky Mountains, Central Asia, equatorial Africa-shrank over the twentieth century. The U.S. Geological Survey and the U.S. National Snow and Ice Data Center have collected air and ground photography of glaciers in the United States showing the extent of glaciers at various times in the past.46 In Glacier National Park in Montana, the melt-off has been dramatic-of the 150 glaciers present in 1850, fewer than 30 are still present today. At the present rate of melting, none will survive past 2030. In Glacier National Park in Montana, the melt-off has been dramatic-of the 150 glaciers present in 1850, fewer than 30 are still present today. At the present rate of melting, none will survive past 2030.

Mount Kilimanjaro sits just a few degrees south of the equator, in East Africa. The equator is an unlikely place to find natural ice, unless you go very high. Kilimanjaro reaches more than nineteen thousand feet above sea level, and for as long as anyone can remember it has had snow and ice at its peak. This iconic image of Africa was immortalized in Ernest Hemingway's short story "The Snows of Kilimanjaro." But throughout the twentieth century, Kilimanjaro has lost ice steadily. The volume of ice present in 2008 was less than 10 percent of what it was a century ago-and at the present rate of loss, ice will disappear from equatorial Africa by 2020.

The Athabasca Glacier in the Canadian Rockies of Alberta is perhaps the most visited glacier in North America, by virtue of its position between Banff and Jasper national parks, two of Canada's favorite scenic treasures. The recession of this glacier is well marked by a succession of signposts installed over the years at the snout of the glacier, which show the glacier's extent at various times in the past. Over the past 125 years the Athabasca has receded almost a mile from the first signpost.

The snowfields and glaciers of the European Alps are also shrinking rapidly, so rapidly in fact that the tourist industry is resorting to desperate measures to slow summertime melting, including laying reflective sheets over the glaciers, as a giant seasonal sunscreen. At the present rate of melting, Alpine glaciers will be only memories by the end of this century. In Asia the glaciers in the Himalayas each year are losing ice equivalent to the entire annual flow of the Huang He, China's fabled Yellow River.

TUNDRA TRAVEL DAYS.

Getting around in the Arctic terrain is never simple, but it is easier, ironically, in winter than in summer. To be sure, the unending daylight of summer offers visibility of unimagined scale, and ease of navigation in an area with few human landmarks. But the broad vistas disguise the fact that in the summer, the ground becomes soft and spongy, depriving vehicles of a firm surface to traverse. The permafrost, the terrain that experiences an average annual temperature below the freezing point, undergoes some limited summertime melting in what is known as the "active zone." This zone extends downward a foot or two or three, and turns a frozen-hard wintertime surface into summertime mush. Off road traffic (and there are very few roads) becomes impossible. Thus, overland transport of supplies to mineral and petroleum exploration camps, scientific stations, and remote settlements is, of necessity, confined mostly to winter.

The time of year when such transport can take place is known as the tundra travel season, and is measured in terms of the number of days that vehicle pa.s.sage overland is possible. Tundra travel days are rapidly diminishing in number. In 1970 one could roll over the frozen surface of northern Alaska more than seven months of the year, but today such travel is possible during only four months, from early January until about mid-May. The overland travel window is closing at a rate of about one month per decade. The tundra surface is now an "active zone" two thirds of the year, and in another half century it may be impa.s.sable year-round.

GREENLAND.

Greenland is an Arctic island bigger than Mexico. It sits almost completely north of the Arctic Circle on the North American side of the Atlantic Ocean. It is a huge reservoir of ice, in volume second only to that of Antarctica. The ice on this large island, covering all but its coastal fringe, is equivalent to more than twenty feet of sea-level change, were it to return to the sea. The top of the ice pile is about twelve thousand feet above sea level, with another thousand feet below sea level because the ice load has depressed the rocky surface beneath it. Greenland's ice slowly creeps downward, and spills into the sea in hundreds of glacial streams around the periphery of the island.

The glaciers are like small holes around the base of a rain barrel-some water escapes through each hole, and in the absence of precipitation, the water level in the barrel will slowly decline. When precipitation into the barrel equals the water losses through the holes, the water level in the barrel remains unchanged, and when the rainfall exceeds the losses at the bottom, the water level will go up. Were there no replenishment of ice in the interior from snowfall, Greenland would eventually be drained of ice.

Every year Greenland undergoes summertime melting around the perimeter of the ice sheet, where the seasonal temperatures at low elevations are sufficiently warm. This band of melting on the fringes has been more or less stable in areal extent and in elevation throughout most of the twentieth century, but toward the end of the century the zone of melting began to creep to higher elevations and over larger areas. The fraction of Greenland's area that undergoes summer melting is 30 percent greater today than it was only thirty years ago, and now ice melts at elevations greater than six thousand feet above sea level.

In midsummer the melting areas are dotted with melt.w.a.ter pools and lakes, beautiful blue jewels accenting the white backdrop. Some of these bodies of water will refreeze in winter, and thus do not represent a net loss of ice ma.s.s. Others lose their water in streams that run to the sea, and these do represent a net loss of ice ma.s.s and contribute to a rising sea level.

It is not an easy task to determine whether the ice budget of Greenland or Antarctica is in surplus or deficit. One technique that has been employed is called repeat-pa.s.s airborne laser altimetry, a method in which an aircraft flies over the ice surface at a low but steady alt.i.tude and repeatedly flashes a laser beam at the surface below. The beam is reflected from the surface back to the aircraft. The time it takes for the laser beam to go down to the ice and return can be translated into the elevation of the ice surface.47 A repeat of the measurement in a few months will show what changes have taken place in the elevation of the ice surface. If the surface has become lower, there is a deficit, and if it is higher, there has been an acc.u.mulation. But elevation changes can be misleading-a fresh snowfall might add three feet in elevation, but because the new snow is light and fluffy, it doesn't represent the same ma.s.s as three feet of dense glacial ice. A repeat of the measurement in a few months will show what changes have taken place in the elevation of the ice surface. If the surface has become lower, there is a deficit, and if it is higher, there has been an acc.u.mulation. But elevation changes can be misleading-a fresh snowfall might add three feet in elevation, but because the new snow is light and fluffy, it doesn't represent the same ma.s.s as three feet of dense glacial ice.

Another technique to determine if an ice budget is changing makes use of detailed measurements of Earth's gravity, as felt by scientific satellites as they orbit the planet. Very small changes in gravity are a.s.sociated with the different densities of the various rocks that make up Earth's crust. Compared to the average, a low-density rock has a ma.s.s "deficiency" and a high-density rock has a ma.s.s "excess." The force of gravity increases over areas of excess ma.s.s and decreases over regions of ma.s.s deficiency. The paths of Earth-orbiting satellites are perturbed very slightly-sped up or slowed down a tad-by these small variations in local ma.s.s and gravity. Thus careful observations of satellite orbits can over time reveal whether a region is losing or gaining ice. A special satellite experiment known as GRACE (Gravity Recovery and Climate Experiment) has been operating since 2002, paying special attention to Greenland and Antarctica. In Greenland, GRACE determined that there is an ongoing ice ma.s.s loss tied to an acceleration of the glaciers draining the interior. The ice deficit for Antarctica has also increased, by 75 percent over the past several years, princ.i.p.ally because of accelerating glacial flow following the disintegration of floating ice shelves around the continent.

AN OCEAN OF ICE.

The Arctic Ocean is a roughly circular ocean with the geographic North Pole at its center. The diameter of the ocean is about 2,800 miles, with North America and Greenland sitting on one side, and Europe and Asia on the other. The entire ocean lies north of the Arctic Circle, and thus experiences the annual extremes of solar illumination-including some days of around-the-clock darkness in winter and unending daylight in summer. For as long as people have been paying attention, much of the ocean has remained frozen year-round in a vast sheet of sea ice. In summer, some of the sea ice breaks up and melts to expose open water, but in winter it refreezes, in a layer about three to six feet thick. During the first half of the twentieth century, about one third of the sea ice melted and refroze each year, leaving two thirds of the ocean with older ice, up to about five years old in places. The older ice is also thicker, occasionally reaching a thickness of fifteen feet or more. Even though much of the Arctic Ocean has been covered with sea ice for at least as long as humans have observed it, the ice is not the same ice. Because sea ice is always on the move-drifting from the Far East, over the North Pole, on toward Scandinavia, and exiting into the Atlantic-no extensive region of the Arctic Ocean has ice much older than five years. The exceptions are in the narrow channels that surround the many islands of the Canadian Arctic, outside of the mainstream of the Arctic drift.

The Age of Exploration-roughly the sixteenth through the nineteenth centuries-coincided with the Little Ice Age cool interval. In the Arctic Ocean, ice formed in every nook and cranny, including in the many channels that wind their way through the archipelago of islands comprising the northern territory of Canada. This maze of waterways, were they to become ice-free, would allow a maritime shortcut from Europe to the trading nations of Asia, a route shorter by two thirds compared to the alternative routes around either Africa or South America. This pa.s.sage, more concept than reality, was called the Northwest Pa.s.sage.

For most of maritime history, however, this route has been closed with ice. Time and again the ice thwarted attempts to open this new trade route to the Orient. On his third and last voyage of discovery aboard HMS Discovery Discovery, Captain James Cook searched for the western entry to the pa.s.sage. Sailing west along the Aleutian Islands in the summer of 1778, he crossed into the Bering Sea near Unalaska Island, and then along the western coast of Alaska to the Bering Strait, with still no hint of a pathway to the east. Northward he continued, through the Bering Strait to lat.i.tude 70 north, where the land began to ease off to the east. Was this the western portal? Cook excitedly began the mental calculations of how long it might take to reach Baffin Bay, that stretch of open water between Canada and Greenland some two thousand miles to the east.

But it was not to be. Two days later Cook saw ice blink, the reflection of a vast expanse of ice on the low clouds in the distance. In a few hours the ice came into view, a solid wall more than ten feet high, as far as the eye could see. Cook recognized the futility of continuing, and turned around to retrace his course back into the Pacific. It would be his last glimpse of ice ever-six months later Captain Cook was dead, killed in a battle with native Polynesians in Hawaii.

Others tried to navigate the Northwest Pa.s.sage from east to west with no more success. The storied Franklin Expedition of 1845-47 became ice-locked about midway through the pa.s.sage, and all aboard perished from starvation. It was not until 1906, when Roald Amundsen, the Norwegian explorer who would five years later gain fame as the first person to reach the South Pole, completed a three-year journey through the Northwest Pa.s.sage to reach Alaska. But the ice he faced may already have been less of an obstacle than that encountered by the eighteenth-and nineteenth-century explorers-the Little Ice Age had reached its peak in the nineteenth century. By 1906, Amundsen was already benefiting from a warming climate.

The summertime retreat and winter refreezing of sea ice are regular cyclical occurrences of long standing, but in the latter decades of the twentieth century, the summer melting began to consume much more than the usual amount of ice, and the winter refreezing fell short of restoring it. By the end of the twentieth century, the summer sea ice had diminished by some 25 percent from its mid-century extent. And as the older ice was replaced by younger ice, the average thickness of the sea ice also diminished, to about half its mid-century measurement.

The Russian icebreaker Yamal Yamal for a number of years has ferried tourists to the North Pole, for a "picnic" on the ice. In August of 2000, everyone aboard was in for a surprise-when for a number of years has ferried tourists to the North Pole, for a "picnic" on the ice. In August of 2000, everyone aboard was in for a surprise-when Yamal Yamal reached the pole, there was only open water. Occasional open water in sea ice is not uncommon-such an ice-free area is called a polynya, a Russian word now in the international lexicon. Polynyas come and go, vagaries of upwelling ocean currents beneath the ice and the wind above. A few polynyas are more or less permanent geographic features, reflecting the stability of the ocean currents, but many others are transient-here today, gone tomorrow. We do not know how common or how rare a North Pole polynya may be, but those who witnessed the occurrence in 2000 remarked that the sea ice had been very thin and peppered with polynyas all the way to the pole. reached the pole, there was only open water. Occasional open water in sea ice is not uncommon-such an ice-free area is called a polynya, a Russian word now in the international lexicon. Polynyas come and go, vagaries of upwelling ocean currents beneath the ice and the wind above. A few polynyas are more or less permanent geographic features, reflecting the stability of the ocean currents, but many others are transient-here today, gone tomorrow. We do not know how common or how rare a North Pole polynya may be, but those who witnessed the occurrence in 2000 remarked that the sea ice had been very thin and peppered with polynyas all the way to the pole. Yamal Yamal's captain said that in all the years he had been traveling to the pole, a polynya there was a first for him.

THE NORTHERN HIGH lat.i.tudes are not unique in signaling a changing climate. In the south, all around Antarctica, the ice is also growing restless. Big ice is the norm in Antarctica, and after coming to the white continent for eighteen years, my jaw does not drop easily. Yet in mid-December 2007, I was awestruck with what was unfolding on the horizon. We were at lat.i.tude 61 south, longitude 54 west, between Elephant and Clarence islands, at the tip of the Antarctic Peninsula, when the biggest piece of floating ice I had ever seen came into view. Thirty-one miles long, twelve miles wide, edged with sheer ice cliffs reaching more than a hundred feet above the sea surface, and another eight or nine hundred feet below-a ma.s.sive island of ice adrift in the southern ocean.

This great slab may have broken away from the Filchner Ice Shelf deep in the Weddell Sea, or perhaps was a fragment of an even bigger ma.s.s that had separated from the Ross Ice Shelf some 2,500 miles south of New Zealand in 2001. Numbers cannot fully describe this floating behemoth. Sixty cubic miles of ice? Fifteen times the area of Manhat tan? The volume of water in Lake Erie? There it was, gigantic ice adrift on a journey to nowhere, pushed along by wind and currents at about two miles per hour.

Sailing alongside this floating ice island I found it impossible to capture the scale-a photo simply showed a cliff extending from the foreground to the horizon. One needs to step back-way back-to be able to see this slab in its entirety. Actually, one needs to step up about a hundred miles, to the viewpoint of an Earth-orbiting satellite, to capture this slab in a single frame. The experience is not unlike feeling the fierce wind of Hurricane Katrina on the ground in southern Louisiana, but needing a satellite image of a giant spinning pinwheel covering the entire Gulf of Mexico to see the full scale of nature's atmospheric fury.

The breaking away of ice of this magnitude from the outer edge of an Antarctic ice shelf certainly begs for attention, particularly when it is not an isolated phenomenon. The Ross Ice Shelf, about the size of France, is the biggest of the huge floating ice sheets nestling along the margins of Antarctica. Others abut both sides of the Antarctic Peninsula, the long and narrow finger-like mountain chain that stretches toward South America. Along the peninsula, mountain glaciers drain ice from the high places, sending it to the sea, where it floats in giant sheets that extend tens and hundreds of miles away from the rocky coast. The La.r.s.en, the Filchner, the Ronne, and the Wilkins ice shelves-named for whalers, scientists, and explorers of a century ago-also are showing wear and tear today.

In early 1996, when I was working aboard MS Explorer Explorer (the expedition ship that sank not far from Elephant Island a decade later), the captain and expedition leader revealed to the expedition staff that we were going to attempt the first-ever circ.u.mnavigation of James Ross Island, named for Sir James Clark Ross, a British explorer who navigated the region in 1842 (and for whom the Ross Ice Shelf is also named). James Ross Island lies near the tip of the peninsula, on the east side; it is the eleventh largest of the myriad islands that dot the fringe of Antarctica. It had been bound tightly by ice at least since any human had viewed it. Yet, there were hints that a circ.u.mnavigation might be possible. (the expedition ship that sank not far from Elephant Island a decade later), the captain and expedition leader revealed to the expedition staff that we were going to attempt the first-ever circ.u.mnavigation of James Ross Island, named for Sir James Clark Ross, a British explorer who navigated the region in 1842 (and for whom the Ross Ice Shelf is also named). James Ross Island lies near the tip of the peninsula, on the east side; it is the eleventh largest of the myriad islands that dot the fringe of Antarctica. It had been bound tightly by ice at least since any human had viewed it. Yet, there were hints that a circ.u.mnavigation might be possible.

Just to the south lay the La.r.s.en Ice Shelf, one of the big ice shelves attached to the peninsula. Two years earlier, the northernmost segment of the La.r.s.en, an area about the size of Luxembourg, had disintegrated, flushing great icebergs into the adjacent Weddell Sea. Could the icy handcuff holding James Ross Island also be loosening? We thought it was a possibility, and began our push into the ice. We were not to be rewarded, however, because the ice was not ready to yield its grip-but the very next year, Explorer Explorer succeeded making it around James Ross Island, through channels that had not seen open water for thousands of years. succeeded making it around James Ross Island, through channels that had not seen open water for thousands of years.

Only five years later an even larger segment of the La.r.s.en disintegrated, one as large as Rhode Island, in a spectacular one-month breakup that delivered so much floating ice to the region that ship navigation was substantially impeded and only ships with a scientific mission ventured into the area. And in late March of 2008 the Wilkins Ice Shelf, on the southwest side of the peninsula, an area about half the size of Scotland, began to disintegrate, shedding floating ice islands of Brobdingnagian scale into the sea. The initial "sliver" from the edge of the shelf was 25 miles long and 1 mile wide, and once it was separated, another 150 square miles behind it quickly broke up. New fractures in the remaining shelf appeared in a November 2008 photo, indicating that the breakup was still progressing, and by April 2009 the disintegration was complete.

The ice shelves around Antarctica are great sheets of ice that have ponded up around the mouths of glaciers that drain ice from the interior. Parts of the shelves may be grounded, but much of the ice is floating as large sheets on the sea. These ma.s.sive, partially anch.o.r.ed shelves serve as b.u.t.tresses that slow the outflow of the glaciers that nourish them-but when the shelves disintegrate, the glaciers find new freedom and speed up their delivery of ice to the sea. A recent survey48 of all the outlet glaciers around Antarctica shows little net loss of ice from East Antarctica (the bigger fraction of Antarctica, east of the Transantarctic Mountains), but substantial and increasing ice loss from West Antarctica and the Antarctic Peninsula. of all the outlet glaciers around Antarctica shows little net loss of ice from East Antarctica (the bigger fraction of Antarctica, east of the Transantarctic Mountains), but substantial and increasing ice loss from West Antarctica and the Antarctic Peninsula.

WHAT DOES THIS accelerating ice loss from both Greenland and Antarctica mean? In a bathtub, the volume of water determines how high the water reaches, and the same holds true for Earth's great natural bathtub-the ocean basins. The volume of water in the oceans rises and falls during the comings and goings of ice ages, and these hydrological transfers are accompanied by changes in sea level of several hundred feet. But in the period of general warmth and relative stability that we have experienced over the past ten thousand years, we have not seen dramatic changes in sea level. There has been a rough equilibrium between losses from the oceans through evaporation, and returns to the oceans via precipitation and the flow of rivers and glaciers to the sea. Over the past several millennia these withdrawals and deposits have continued to take place in the oceanic account, but the balance has remained pretty steady.

But the twentieth-century warming of Earth and the loss of ice from the continents are beginning to change the oceanic balance-in the upward direction. Two princ.i.p.al factors are at work. The first is the increased melting of ice and the return of the melt.w.a.ter to the sea, or alternatively the direct deposit of ice into the sea from faster-flowing glaciers. The second factor is the volumetric expansion of seawater as the oceans warm. The volume of most liquids increases when the tempera-ture goes up-that is the fundamental principle behind liquid-in-gla.s.s thermometers that show the level of the liquid rising in the scaled gla.s.s tube as the temperature rises.

Changes of water level during a flood episode along a river are apparent in vertical changes-how high the water rises along a levee or the wall of a building-and in horizontal changes seen in the growth in the area covered by water. Sea level changes are apparent in these same ways. The measurement of the vertical change, the amount of sea-level rise, is accomplished by an instrument called a tide gauge, the primary purpose of which is to measure the amplitude of the high and low tides in a bay or harbor. This instrument records the changing water level during the daily rise and fall of the tide, but over years, decades, and centuries it also shows the long-term changes a.s.sociated with slowly rising sea level. Data from thousands of tide gauges the world over have been collected and a.n.a.lyzed, and show that during the twentieth century, sea level on Earth rose about eight inches. One third of the rise comes from new deposits of melt.w.a.ter and ice into the sea, and two thirds from the thermal expansion of the warming oceans.

But it is the horizontal incursion of the rising sea that is most apparent to the eye. On a gently sloping beach, a small rise of sea level will extend quite a ways inland to define a new sh.o.r.eline. Eight inches of sea-level rise on a beach with a gentle slope of one degree will move the sh.o.r.eline almost forty feet inland. And the same forty feet will be subject to the daily flooding of the high tide, and more vulnerable to the high-water surges from storms at sea.

THE IPCC.

1n 1988 the United Nations created an international scientific working group called the Intergovernmental Panel on Climate Change (IPCC). The charge to this group was to a.s.sess whether climate change was occurring, what the causes of such climate change might be, what the consequences of past and future climate change have been and might be, and what options might exist for mitigation of, and/or adaptation to, a changing climate on Earth. The task of organizing this panel fell to the World Meteorological Organization and the United Nations Environment Programme, both ent.i.ties of the United Nations.

The IPCC is not a research organization itself, but rather an evaluator and summarizer of peer-reviewed scientific research published in scholarly journals the world over. It issues periodic a.s.sessment reports every five years or so, describing the state of knowledge about climate change. These a.s.sessment reports have appeared in 1990, 1995, 2001, and, most recently, 2007. Components of the reports are written by teams of scientists active in the various subfields of climate science, then collected into chapters by a group of "lead authors"; the chapters are a.s.sembled into a coherent and seamless a.s.sessment by an even smaller number of "coordinating lead authors." Altogether, more than two thousand active climate scientists contributed to the Fourth a.s.sessment Report published in 2007.

A few words about the process of peer review in the IPCC a.s.sessments: peer review is essentially a process of quality control in the world of serious scholarship. Publication of research results in a peer-reviewed journal means that an article has been read by other practicing researchers in the area, and a.s.sessed for originality, appropriate methodology, data quality, and sound conclusions. Most articles that appear in journals have been revised once or twice prior to publication in response to critiques from the reviewers. Submitted articles that fail the review are, of course, rejected.49 The published scientific results considered by the IPCC have already been peer-reviewed by the independent journals in which they were published, but that is not the end of peer review in the IPCC a.s.sessment reports. After a draft of an a.s.sessment report has been prepared, it is sent to a large pool of climate scientists not involved in its writing, but active in and knowledgeable about the fundamental science. They are asked to provide another layer of peer review to determine whether the draft a.s.sessment report is accurate, balanced, and free of distortion or exaggeration. Critiques can be formally expressed, and then forwarded to the a.s.sessment authors for a response. The authors are obliged to respond to each critique, either accepting and incorporating or rejecting and reb.u.t.ting the essence of each commentary. More than thirty thousand written comments were submitted by more than six hundred individual expert reviewers of the Fourth a.s.sessment Report's volume on the physical science of climate change.

The revised a.s.sessment report is next forwarded for review to the governments of the member countries represented in the United Nations. At this level the issues discussed are a blend of science, economics, and policy, but ultimately the language of the a.s.sessment reports must be approved by the governments. Considerable debate, accompanied by nuanced word-crafting, takes place, sometimes requiring an "agreement to disagree," but ultimately the text is approved and the a.s.sessment report becomes official and publicly available. For the Fourth a.s.sessment Report in 2007 some 130 governments partic.i.p.ated in this final stage of review.

The reason I have gone to great length to describe this review process is to make clear that, in the end, the IPCC report is a doc.u.ment that must, by any measure, be deemed conservative. The review process weeds out unbounded speculation, problematic science, and untested hypotheses. It carefully evaluates and states the uncertainties at every step of the way. In the end, what results is a lowest-common-denominator consensus of what the science is telling us. Moreover, the resulting reports are not policy prescriptive: that is, they do not tell governments what to do. They simply lay out various scenarios with attendant consequences: if you do X, you can expect Y; if you don't do W, you can expect Z.

The thorough and systematic quality control exercised by the IPCC contrasts strongly with the communications of the climate contras. These "skeptics" choose newspapers, radio, and television for their "scientific" p.r.o.nouncements, or publications subsidized by vested interests that want to discredit what the real science is revealing. The contras have little interest in persuading the mainstream scientific community, and care little about peer review. The audiences they aim to persuade are state legislators and members of Congress, and other governing bodies around the world, where climate policy will ultimately be shaped.

THE VERDICT: "UNEQUIVOCAL"

So what did the IPCC's Fourth a.s.sessment Report say about the evidence that Earth's climate is changing? Here is its bottom line: Warming of the climate system is unequivocal, as is now evident from observations of increases in global average air and ocean temperatures, widespread melting of snow and ice, and rising global average sea level.50 Ice everywhere is talking to us-not politically or emotionally or conventionally-but in a language that we must understand and heed. Ice is a sleeping giant that has been awakened, and if we fail to recognize what has been unleashed, it will be at our peril.

The IPPC's use of the word unequivocal unequivocal leaves little wiggle room. It means there can be no confusion about it. There can be no mistake about it. "Maybe, maybe not" is over. Significant climate change is happening. Seldom do we hear scientists make such an unambiguous p.r.o.nouncement. leaves little wiggle room. It means there can be no confusion about it. There can be no mistake about it. "Maybe, maybe not" is over. Significant climate change is happening. Seldom do we hear scientists make such an unambiguous p.r.o.nouncement.

It is time to move on to other issues. Let us turn now to the causes of climate change. In the next two chapters we look first at the natural factors and then at the human factors that can cause Earth's climate to change on the time scale of the twentieth-century warming.

CHAPTER 5.

NATURE AT WORK.

The bright sun was extinguish'd, and the stars Did wander darkling in the eternal s.p.a.ce, Rayless, and pathless, and the icy earth Swung blind and blackening in the moonless air; Morn came and went-and came, and brought no day . . .

-LORD BYRON "Darkness" (1816)

In early April of 1815 the city of Batavia, on the island of Java, began to hear sharp explosions that sounded much like distant artillery. But the city was not under attack, and no ship in distress was firing its cannons as a call for rescue. Batavia (now Jakarta) would later learn that it had been an aural witness to a violent eruption of the volcano Tambora, 1,200 miles to the east, along the Indonesian archipelago. For over a week Tambora erupted in a series of explosive events, creating sound waves heard even another 800 miles beyond Batavia. More than 70,000 people perished from the triple plague of a red-hot noxious gas and debris cloud rolling down the mountainside, a tsunami generated by the culminating explosion of April 11, and contamination of drinking water by the prolific ashfall. But if the devastating loss of life nearby was the immediate consequence, the eruption would continue to cause problems worldwide in the months and years to follow.

The explosions reduced the summit of the mountain from around 14,000 feet to 9,500 feet, blowing about 35 cubic miles of the volcanic edifice-an astounding volume of debris-into the atmosphere. In just a few months the atmospheric jet streams distributed this debris worldwide, blocking some of the sunshine from reaching and warming Earth's surface. The atmosphere is not quickly purged of such a burden, and the climatic effects were very apparent the following year. Temperatures throughout the Northern Hemisphere were depressed well below normal, and crop failures were common in Europe and North America. Connecticut experienced snow in early June, lakes in Maine froze over in mid-July, the mountains of Vermont were snow-covered in August.51 Four killing frosts-one in June, one in July, and two in August-ensured that the New England harvest was meager. Around the world, the year 1816 became known as "the year without a summer." Four killing frosts-one in June, one in July, and two in August-ensured that the New England harvest was meager. Around the world, the year 1816 became known as "the year without a summer."

Ash, dust, and chemical aerosols injected into the atmosphere during volcanic eruptions form a veil that blocks some of the incoming sunshine and prevents it from reaching and warming Earth's surface. The volcanic products have the effect of increasing Earth's albedo for a few years, reflecting more incoming solar energy back to s.p.a.ce. Eventually, the ash and dust fall back to Earth, clearing the atmosphere and allowing the Sun's rays to once again warm the planet.

The eruption of Tambora was neither the first nor the last volcanic event to have a global impact on the atmosphere and climate. The diminished sunshine following an eruption in AD 536 on the island of New Britain, just east of Papua New Guinea, led to this description of conditions in the Middle East: The Sun became dark and its darkness lasted for eighteen months. Each day it shone for about four hours, and still this light was only a feeble shadow. Everyone declared that the Sun would never recover its full light. The fruits did not ripen and the wine tasted like sour grapes.52 The dust from this eruption has been detected in well-dated ice cores in both Greenland and Antarctica. Tree-ring data from the European Alps, Scandinavia, and the Russian Arctic suggest that the cooling caused by this eruption may have been the most severe that the Northern Hemisphere has experienced in the last two millennia, even cooler than the effects from the 1815 eruption of Tambora.53 In August of 1883, another Indonesian volcano, Krakatoa, in one mighty explosion spewed more than six cubic miles of ash and dust into the atmosphere. The sound of the explosion was heard in Mauritius, an island located in the deep south of the Indian Ocean, some three thousand miles away, and the particles injected into the atmosphere soon led to spectacular red sunsets around the world. The colorful skies were captured in a series of sketches by the English painter William Ashcroft in 1884, and became known as the Chelsea sunsets. The official scientific report on the eruption of Krakatoa, published by Britain's Royal Society in 1888, featured the Ashcroft sketches as its frontispiece. For years following the eruption, brilliant sunsets and brutal winters were experienced around the world. The legendary harsh winter of 1886-87 and the devastating blizzards of 1888-Krakatoa's unwelcome gifts to the struggling settlers and ranchers in the Great Plains-brought cattle-grazing on the open range of the United States to an end.

These examples make clear that great volumes of volcanic debris sent high into the atmosphere during an eruption, and soon thereafter distributed around the globe by the atmospheric circulation, can affect the global climate for several years. Volcanism is but one of the arrows in nature's quiver of climate-changing mechanisms that have played a role in Earth's climate in the aeons before humans populated the globe. Let us now look at some of the other processes that have altered Earth's climate prior to the appearance of humans.

WHEN THE CLIMATE is not changing, there is an equilibrium between the incoming energy absorbed by Earth's surface and the outgoing energy radiated back to s.p.a.ce from the surface. All factors leading to climate change disrupt this balance between energy deposits and withdrawals from Earth's surface. Disruptions to the equilibrium include changes in the amount of sunshine arriving from the Sun, changes in the fraction of that energy that Earth reflects back to s.p.a.ce, and changes in the atmosphere that cause it to capture some of Earth's heat instead of allowing it to radiate back to s.p.a.ce unimpeded.

THE SUN DELIVERS.

The amount of radiant energy Earth receives from the Sun changes over time, and not just because of variations in the amount that leaves the Sun. The periodic changes in the ellipticity of Earth's...o...b..t around the Sun and in the tilt and precession of Earth's rotational axis-the Milankovitch cycles described in chapter 3-affect both the distance of Earth from the Sun and how Earth is oriented with respect to the Sun. Being closer to or farther from the Sun will have obvious effects on Earth's temperature, and the changes in the tilt and orientation of Earth's rotational axis affect the strength of the seasonal variation in temperature.

But these Milankovitch cycles, with seemingly long periods of one hundred thousand, forty-one thousand, and twenty-three thousand years, are mere flutters on a very slow increase in solar luminosity that has been occurring since the beginning of our solar system four and a half billion years ago. This long-term increase in the Sun's radiance is a common evolutionary characteristic of millions of stars similar to the Sun. At the birth of the solar system the primeval Sun displayed only 70 percent of the luminosity it displays today. A dimmer Sun in that very early history of our solar system would imply that Earth was much colder in its early days, and ice much more common. Calculations of Earth's surface temperature with only 70 percent of today's solar radiation warming its surface inevitably translate into an ice-covered early Earth. And yet, geologists have identified widespread sedimentary rocks-rocks deposited in water-that are almost as old as Earth itself, suggesting that liquid H2O was present early in Earth's history. This apparent contradiction was named the "faint young Sun paradox" by astronomers Carl Sagan and George Mullen.54 The paradox can be resolved with an atmosphere that also evolved in response to the slowly increasing solar radiation. Although nitrogen has probably always been the princ.i.p.al chemical component of Earth's atmosphere, oxygen has not. Oxygen in the atmosphere today is the princ.i.p.al waste product of photosynthesis, the process by which plants use sunlight to produce bioma.s.s-in other words, to grow. But photosynthesis did not become an important source of oxygen in the atmosphere until green plants evolved later in Earth's history. Initially there was very little oxygen in the atmosphere, because, absent photosynthesis, the only other process producing oxygen was a weak mechanism called photodissociation, in which radiative energy from the Sun broke the chemical bonds between the oxygen and hydrogen atoms in some of the water vapor molecules in the early atmosphere, thus freeing up a little oxygen. Even today, photodissociation yields much less oxygen than does photosynthesis, and under conditions of a dimmer Sun early in Earth's history, it would have been an even less efficient process.

With little oxygen available, carbon in the early atmosphere joined hydrogen to form methane, CH4, in the atmosphere. Methane, however, is a potent heat-trapping gas, some twenty times stronger in its heat-trapping capability than its oxidized cousin carbon dioxide, CO2. Earth's early atmosphere therefore acted as an extraordinarily effective blanket. As oxygen slowly became more abundant over geologic history, carbon dioxide gradually replaced methane, and the heat-trapping ability of the atmosphere slowly declined. When the Sun was weak, Earth's atmospheric heat-trapping blanket was strong, and as the Sun grew more radiant, the blanket grew weaker. The result is that Earth's average surface temperature has remained in the range of liquid H2O throughout most of its history. The gradual oxidation of the atmosphere is now recognized as the resolution of the "faint young Sun paradox."

EARTH AS A GREENHOUSE.

However much energy the Sun delivers to Earth, that energy can be diminished or enhanced by processes within our atmosphere. Explosive volcanism (as described earlier in this chapter) can block some of the solar radiation from reaching Earth's surface, and greenhouse gases in the atmosphere impede the escape of Earth's heat back to s.p.a.ce.

How do the greenhouse gases trap heat trying to leave Earth? For a decent a.n.a.logy, think of the microwave oven in your kitchen, in which microwaves (electromagnetic waves larger than infared but shorter than radio waves) are generated within the oven and absorbed by the food you want to heat. More specifically, the microwaves are absorbed by water molecules contained in the targeted morsels. Because the microwaves are absorbed within the food, the energy they carried as traveling waves is converted to another form of energy, heat, as required by the first law of thermodynamics.

Let's now scale the concept up to solar system dimensions to get a feel for Earth's natural (and long-standing) greenhouse effect. The dominant radiation the Sun generates is in the visible part of the electromagnetic spectrum, that band of wavelengths that our eyes have evolved to be sensitive to. These wavelengths include all the colors of the rainbow-red, orange, yellow, green, blue, and violet. The Sun sends off a little energy outside of the visible range-some shorter ultraviolet waves and some longer infrared waves-but most of the energy arrives in the visible wavelengths.55 Our atmosphere is essentially transparent to the visible wavelengths, so this energy from the Sun pa.s.ses through the atmosphere unimpeded, to be absorbed at Earth's surface and warm it. But Earth cannot continually absorb energy and keep on heating up forever, at least not without serious consequences, such as melting. It must have a way of sending heat back into s.p.a.ce to avoid continual warming. It accomplishes this balancing act by reradiating the energy received from the Sun back to s.p.a.ce, but not in the visible wavelengths. The wavelengths that a body employs to radiate energy away depend on the temperature of the surface, with hotter bodies such as the Sun radiating shorter waves, and cooler bodies such as the planets radiating longer waves.

Energy comes to Earth as visible radiation from a very hot (~11,000 Fahrenheit) Sun, but departs as invisible infrared radiation from a 60 Fahrenheit Earth. But now comes the hooker-the atmosphere, which is transparent to incoming visible radiation, is not fully transparent to the outgoing infrared waves. Several gases in our atmosphere, present in only tiny amounts, absorb infrared radiation and convert the radiant energy into heat. This is what we call the "greenhouse effect"-the process by which Earth takes in a little more heat than it sends back, and accordingly it must warm up a bit and radiate a little more, in order to restore the balance between incoming and outgoing energy.

The greenhouse effect is not simply some theoretical scientific construct-it is a very real observable and measurable phenomenon, and one we should be thankful for, because Earth would be much colder and inhospitable without it. The princ.i.p.al gases in the atmosphere that absorb infrared radiation are water vapor (H2O), carbon dioxide (CO2), and methane (CH4); together they add up to less than 1 percent of the atmosphere. For every million units of atmospheric volume, only a few hundred parts are CO2, and less than two parts are CH4-but these minuscule amounts give a lot of "bang for the buck." One sometimes hears incredulity that such tiny concentrations can have any impact, let alone a major one. But these trace gases in our atmosphere raise Earth's surface temperature by more than sixty Fahrenheit degrees from what the surface temperature would be if Earth had no atmosphere. This natural greenhouse effect is what makes Earth the water planet, the blue planet, rather than just another of the many icy bodies of the solar system.

The absorption of infrared radiation by CO2-an atmospheric process that would become so discussed in the second half of the twentieth century-was first measured by John Tyndall in 1859. It is no small historic irony that 1859 was the same year that petroleum was discovered in Pennsylvania by Edwin Drake. Little did anyone imagine that a century later the CO2 from the combustion of petroleum would be warming the atmosphere by the mechanism first measured by Tyndall. from the combustion of petroleum would be warming the atmosphere by the mechanism first measured by Tyndall.

EARTH'S GEOLOGICAL THERMOSTAT Earth has a "thermostat" that prevents the surface temperature from straying too widely. It does not, however, make adjustments daily, as in our homes. Rather, the adjustments take place over millions of years and are related to geologic processes that are temperature dependent. The conceptualization of this thermostat originated with Jim Walker, a very broad-based earth scientist at the University of Michigan. Walker synthesized perspectives from atmospheric science, oceanography, geology, and geochemistry to envision the way this geological thermostat works.56 Imagine an Earth that is a little too warm because of an atmosphere with an above-average concentration of the greenhouse gas CO2. How does Earth turn down the thermostat? Some chemical reactions that decompose rock-a process that geologists call weathering-are more effective at higher temperatures, and so when Earth is warmer, the rivers that drain the continents carry a bigger load of dissolved chemicals to the sea.

One element that weathers from continental rocks is calcium, which, when delivered to the sea, combines with carbon dissolved in seawater to produce calcium carbonate, which ultimately is deposited on the seafloor as limestone. As carbon is removed from the seawater through limestone deposition, the sea pulls more CO2 from the atmosphere, thus diminishing the greenhouse effect and cooling the planet. But as the surface cools, less weathering takes place, the supply of calcium to the sea slows, limestone deposition diminishes, and once again CO from the atmosphere, thus diminishing the greenhouse effect and cooling the planet. But as the surface cools, less weathering takes place, the supply of calcium to the sea slows, limestone deposition diminishes, and once again CO2 builds up in the atmosphere to warm the planet.. builds up in the atmosphere to warm the planet..

Earth's temperature oscillates between warmer and cooler through fluctuations in the effectiveness of the natural atmospheric greenhouse, which in turn modulates the availability of calcium to form limestone. But the functioning of this thermostat is dependent on oceans full of water in which to absorb CO2 from the atmosphere and deposit limestone on the ocean floor. Without water, Earth's thermostat would be broken. from the atmosphere and deposit limestone on the ocean floor. Without water, Earth's thermostat would be broken.

VENUS-A PLANET WITHOUT A THERMOSTAT Earth's closest neighbor in the solar system is Venus, the second rock from the Sun. Venus is similar to Earth in many ways: it is about the same size, its gravity field is about the same strength, its chemical composition parallels that of Earth, and it has an atmosphere. Its greater proximity to the Sun, at about only three quarters of Earth's distance from the Sun, would suggest a surface temperature warmer than Earth's, but still in the range that, if there were any H2O present, it would be liquid water rather than ice or water vapor. So it was quite a surprise when planetary scientists discovered that the surface temperature of Venus was in excess of 860 Fahrenheit. This temperature is high enough to melt lead, and way too warm for water to exist at the surface of the planet.

What has happened on Venus? The clues emerge from the composition and ma.s.s of its atmosphere-a gaseous envelope around the planet nearly one hundred times more ma.s.sive than Earth's atmosphere, and composed almost entirely of CO2. In short, Venus has a thick greenhouse blanket that has trapped enough heat to raise the planet's surface temperature more than eight hundred Fahrenheit degrees higher than it would have without such an atmosphere. Comparatively, both Venus and Earth have a similar amount of carbon, but on Earth only a tiny fraction of the carbon is in the atmosphere. Most of Earth's carbon resides in deposits of coal, petroleum, natural gas, and limestone. In other words, Earth has stored most of its carbon underground, whereas carbon on Venus resides almost entirely in its atmosphere. What a difference that makes in the surface temperature!

Why is Venus unable to regulate its surface temperature in the way that Earth's geological thermostat has maintained a water-compatible temperature on Earth? The likely reason is that Venus is closer to the Sun, at a distance where it receives almost twice as much solar energy as Earth does. There, more intense evaporation led to a complete depletion of surface water, and without water there can be no biosphere to create coal, no oceans in which to deposit limestone-in short, Venus had no way to sequester carbon in solid form, and so carbon simply acc.u.mulated as gaseous carbon dioxide in the atmosphere, creating an intensely effective greenhouse blanket.

SHORTER-TERM FLUCTUATIONS IN CLIMATE.

The slow evolution of the Sun over billions of years and the geological thermostat regulating temperature over millions of years cannot explain significant changes in climate over decades or a century. Those long-term natural processes change so slowly that effectively one century looks pretty much like the next. To explain the relatively fast contemporary warming in the twentieth and twenty-first centuries, we need to look for other causes. The variability of the energy radiated from the Sun has always been recognized as an important natural factor driving changes in Earth's climate.

Solar physicists and astronomers have learned from years of research that the Sun is a very active body, and that the amount of radiative energy leaving the Sun varies over many time scales-minute by minute, day by day, year by year, decade by decade. All of these rapid fluctuations ride along on the three long Milankovitch cycles, which create a slowly changing backdrop for the changes in solar output that occur on shorter and less reliably periodic time scales.

Short-term variations in solar output can impact Earth in a variety of ways. Occasional solar flares leaping upward a million miles above the Sun can be so intense that they disrupt radio transmissions on Earth for several days and even damage the electronics of orbiting communication satellites. These solar outbursts are a serious military concern, because they can blind battlefield surveillance from s.p.a.ce and interrupt the remote control of the pilotless drones that are playing an increasingly important role in military operations.

The decadal variability of the solar output can be seen in the abundance of sunspots, dark irregular patches that appear on the face of the Sun. Sunspots have been observed astronomically for more than four hundred years.57 Their numbers wax and wane with an apparent period of about eleven years-give or take a year. However, the strength of each cycle, as indicated by the number of sunspots at a cycle's peak, show upward or downward trends extending over centuries. Their numbers wax and wane with an apparent period of about eleven years-give or take a year. However, the strength of each cycle, as indicated by the number of sunspots at a cycle's peak, show upward or downward trends extending over centuries.

Generally speaking, the more sunspots there are, the more energy the Sun radiates. Conversely, when the Sun shows fewer spots, its radiative output is correspondingly diminished. The dark patches are actually colder than the brighter areas of the Sun that surround them, and so one might imagine that a darker Sun-one with more spots-would be radiating less energy than a brighter Sun. But the real meaning of more spots is that the Sun is churning more vigorously, and bringing more energy to its surface for radiation into s.p.a.ce. When the Sun is "quiet," the spots are few, and outbound radiant energy is less.

In the period 1650-1715 there were very few spots on the face of the Sun, a period known as the Maunder Minimum, named for the nineteenth-century English astronomer Edward Maunder, who first pointed out the paucity of spots. On Earth it also coincided with a particularly cool interval within the broader climate downturn known as the Little Ice Age (mentioned in chapter 4), during which glaciers advanced in their valleys and the growing season grew shorter.

In the first half of the twentieth century, the peaks of the sunspot cycle grew for several cycles in a row. These decades of an increasingly active Sun probably contributed to a climb in the global average temperature of almost 0.9 Fahrenheit degree from 1910 to 1950. But in the last four decades of the twentieth century, the Sun has undergone a modest decline in radiative output. The minimum of the sunspot cycle in 2008 was the lowest in the past half century.

Since 1978, scientific satellites...o...b..ting above our atmosphere have measured the incoming solar radiation in great detail-not only in the visible light of the electromagnetic spectrum, but also in the shorter ultraviolet and longer infrared wavelengths. These observations provide much more comprehensive information about the variability of solar radiation than does the simple counting of sunspots. The essential story the satellite radiometers tell, however, is the same as the sunspots: solar radiation has been declining in the latter decades of the twentieth century. Despite that, Earth's temperature has continued to climb over the same interval. Apparently, the Sun is sharing the stage with other factors that are affecting Earth's climate and causing it to warm.

VOLCANIC DIMMING IN THE TWENTIETH CENTURY.