Reading by the pool? Underrated. Really, the end of summer is an opportune time for looking at recent heat events in a longer-term context, and thinking about what can be learned from using their differences as a window into the broader categorizations of heat waves. 2015, like 2014 before it, is on track to nominally be the overall warmest year in the observational record (due to incomplete spatial sampling, adjustment uncertainties for urbanization and elevation, etc, this can never be completely certain, so several other years have a nonzero probability of actually having been the warmest, all of them since 1998; in other words, of all the years since 1880, 2014 wins first place by a plurality). Of course, as can be seen in the map at the first link, there is always large spatial variability around the globe, so not everyone feels this warmth firsthand; indeed, it does not rule out 2015 being the coldest year at certain sites (chart is for Ithaca, NY, with the black line representing 2015 through Aug 28). But it does, naturally, predispose a greater severity (intensity+duration) of heat waves overall. A recent paper examined this truism with several models to see how projected warming would affect the severity of heat waves over the course of the 21st century in regions that have recently experienced record-setting heat events. One figure from that paper is shown below: using a custom-defined "heat-wave magnitude index" that is essentially the sum of consecutive hot maximum temperatures, a few K of mean warming translate to heat waves every year in the late 21st century much more severe than any observed up to this point. Many other studies using different definitions and models have come to similar conclusions. Some notes of caution are due, however. There are general irreducible uncertainties about the models -- though in this case, with averaging over large regions, these standard (CMIP5) models hew closely to observations when tested over the historical period. There are synoptic kinds of uncertainty, such as whether blocking (stalled pressure systems) turns out to behave somehow differently under similar governing conditions than it does at present. And there is the question of the representativeness of this particular heat-wave magnitude index for human-health impacts and ecosystem vulnerability. Minimum temperature has generally been found more important than the maximum, for instance, and impacts are always a combination of duration, intensity, and frequency in unknown proportion, all interacting with cultural and physiological factors. Interestingly, there has also been some indication that acclimatization on multi-year timescales has already occurred in France (in addition to the perennial seasonal kind discussed in an earlier post), suggesting that this may help to pull down the impacts from the ballooning predicted severity of future heat waves. ![]() Projections for a maximum-temperature-defined "heat-wave magnitude index" over the globe and in several regions with well-studied record-setting events that were large, intense, and long. Red and blue in the observational record indicate different datasets; green, light blue, and gray in the future indicate the RCP 8.5, 4.5, and 2.6 climate-change scenarios. Source: Russo et al. 2014. So what were the major stories of the summer, heat-wise? A blog post by Jeff Masters summarizing the events to date as of July 1 already had a lot to talk about. The India/Pakistan heat wave probably received the most attention, particularly centered around the death toll in Karachi, with similar stories from India as well (though less exacerbated there by the water-forgoing strictures of Ramadan). High temperatures but minus the large cost in lives continued to be observed in parts of western Europe in July and August, with July temperatures in Geneva rivaling those of the infamous August 2003 heat wave. Perhaps most memorably, the Iranian city of Bandar Mahshahr on the Persian Gulf saw its heat index (the version as defined by the National Weather Service) soar to 165 F/74 C, pushed along by an exceptionally high dew point of 90 F/32 C (for a modest-sounding relative humidity of 47%). A Weather Channel graphic from that evening is shown below. These conditions aren't even on the NWS heat-index chart; after all, temperatures almost always can only get that high with relative humidities below 20% (witness Baghdad in the figure) -- because with more moisture in the air, solar heat input goes more to heating water vapor than to raising the temperature. Whether or not the standard formulation of this heat index is a reasonable representation of the physiological stress under those exceptional conditions is debatable (it's not valid under 80 F either, and there were no reports of mass casualties); nevertheless, a purely maximum-temperature-oriented approach would not capture the surely suffocating experience of 115-dewpoint-of-90. These kinds of humid extremes are poorly understood but clearly result from a rare degree of interaction between hot air from the inland deserts and moist air from the Persian Gulf along a coastal strip. In Dhahran, eastern Saudi Arabia, location of a world-record 95 F dewpoint in 2003, summertime maximum temperatures have increased over the last few decades, as have minimum temperatures, although the changes are spatially variable within the Arabian Peninsula. Dewpoint temperatures have modestly decreased (but sharply increased just to the south; see Fig. 15 in last link). Trends contributing to humid extremes: inconclusive. Overall, summer 2015 did see an abnormally high intensity and spatial coverage of extreme heat relative to the observational record, though on longer timescales we can only say probabilistically that with lower mean temperatures in the past we expect there to have been fewer and/or less-severe heat waves. Don't even try to look for trends in extreme heat indices -- there are too many definitions and too few data to come anywhere near a clear conception of that. And for impacts, it is surely not the case that Europe 2003 and Russia 2010 actually stood out as the deadliest, or even deserve a spot on the all-time top 10; like with temperatures themselves, changes in definitions, preparedness, and awareness dominate over the environmental conditions. The more angles we look at heat waves from the fuzzier our simple mental conception of 'high temperatures' gets, but conversely the more accurate it becomes.
0 Comments
As climate has moved more and more into the consciousness of the people and organizations who make political, investment, and development decisions that affect all of society, the vocabulary used to describe the various ways in which such decisions engage with climate or climate projections has increased commensurately. We now have 'impacts' on 'stakeholders' which we address with 'adaptation', 'mitigation', and 'resilience', but also with broader measures of 'sustainability' and reductions in 'vulnerability'. Even as someone who encounters these terms every day, I realized I might be hard-pressed to give an exact definition of them, much less to give a coherent explanation of why in a particular circumstance one would be preferable to another. Only once the terms are all on solid linguistic ground can the ultimate and inherently local question be taken up of when one type of reaction is preferable to another, given a suite of social, economic, and cultural factors. The first of these terms is so common that I placed it in the title of this post without even noticing. The IPCC glossary defines impacts as "the effects of climate change on natural and human systems"; this is a fine definition for an organization that has the CC of climate change right in its name, but from a broader perspective natural climate variability (e.g. ENSO) should be included as well. After all it's the superposition of anthropogenic and natural climate forcings that will determine the future climate that planners must account for. 'Stakeholder' is similarly vaguely defined in terms of one's potential to be affected 'by an entity' -- in this case, the entity being the future climate in the region of the stakeholder's interest. In policy parlance 'stakeholders' and 'decision makers' tend to be strongly overlapping sets; in a typical region they might consist of various municipal-government departments, utilities, corporations, community boards, and the like. The concerns of individual households are assumed to be covered by this array of larger organizations, making it crucial that a wide variety of stakeholders are included in selecting research topics and deliberating on actions -- but not too wide, lest research efforts be spread too thin. This tradeoff follows the precepts of Arrow's impossibility theorem, where having fewer moving parts in a decision-making machine is often desirable despite thus less precisely matching the full distribution of societal preferences.
'Sustainability' is an overloaded station wagon of a word that even in the specific urban-climate context can take on a number of different meanings. Most broadly, it is used to refer to any system or action that reduces the consumption of some natural resource (energy, land, water, materials, etc) and by consequence the ecological footprint of the population. A full evaluation of it would require each person tracing back the resources consumed in the production of every commodity they use, wherever they came from and whatever steps were taken to get them into their final forms. It's typically the reduction in the resources that matters, as it's impossible to say exactly what can be sustained per se, and over what time period the accounting is done. When the flux of resources through our lives is reduced, we say that we are less vulnerable, meaning less susceptible to disruptions. This is a particularly salient point for urban residents the world over whose food, energy, clothes, devices, etc must be shipped in every day, even a brief halt to the system causing weeks-long ripple effects. Vulnerability is often broken down into sensitivity (e.g. the percent chance that someone will suffer heatstroke at a certain temperature) multiplied by exposure (the number of exposures per time period); to obtain the societal impact, multiply again by the population affected. Technological advances and transitions away from agricultural occupations may well reduce global exposure to climate extremes despite increasing numbers of extreme events overall. In this solar system of vocabulary, where newly coined words hurtle around like asteroids, mitigation and adaptation are like the binary stars, at the center of most every conversation. Fittingly, neat definitions of the two are given on NASA's "Vital Signs of the Planet" website. In short, mitigation is mostly about greenhouse gases, while adaptation is about animals, plants, water, non-extreme temperatures, and all the other things that people need and climate affects. A decade or more ago mitigation was the preferred tack, via emissions cuts most prominently, but as the world has continued on a high-emissions pathway (see post "The Human Element: Time Dependence") adaptation has gone from being seen as a concession to something worth striving for -- indeed, that needs to be striven for in the continued absence of serious mitigation efforts. (Adaptation is sometimes rolled into the definition of vulnerability as well.) The talks in Paris this coming December will provide another chance for international reassessment of commitments to each vis-à-vis the other. Recently, talk has moved beyond adaptation to resilience, and specifically the sense of the word suggesting that failure in the face of uncertain and changing conditions is not in and of itself disastrous, and that good preparations that happen are better than great ones that don't. The nuances of resilience are nicely analyzed in this blog post. The term "graceful failure" encapsulates the essence of this new concept, wherein a certain percentage and/or magnitude of failures are deemed acceptable. On the other hand, with distributions of climate variables often being inherently long-tailed, the merits of such an approach are debatable. It only takes one very extreme event to make irrelevant a lot of well-laid plans, so a reasonable case could be made that the absolute worst-case scenario should be at least considered, insofar as is possible. The insurance industry uses statistical models run hundreds of thousands of times to get estimates of the extreme tails of the distribution of whatever variable (wildfires, floods, heat, etc) they are financially exposed to. The burden of this computation makes greater sense at longer time horizons, so it doesn't make sense in every circumstance, but generally speaking knowing the bounds with reasonable certainty is an important prerequisite of resilience, which in turn backstops more-mundane adaptation and mitigation measures so that stakeholders can reduce their vulnerability in the near term while also keeping an eye on climate impacts over much longer time horizons. In a departure from some of the usual more research-laden posts, I will spend some time every once in a while examining topical issues revolving around the theme of city administrative and/or social adaptations to climate, and in particular to extreme climate events. Early July seems like the perfect time to delve into the first question I will look at: how cities deal with snow removal. Rather than being a complete non-sequitur, this was prompted most immediately by a recent article in the New York Times on the last garbage-filled hard-packed remnant of the enormous snow piles created in the wake of Boston's record-setting snowfall last winter. (Which set off the kind of repartee of dismissiveness often characteristic of the discourse between those two cities.) This mound's placement on the docks, but not actually in the water, caused it to reach 75 ft high (and made the mounding process difficult in the days following the storms) but kept the trash within safely out of the harbor. An article inspired by Boston's previous unusually snowy winter discussed several of the actions taken by other cities, including giant snow-melting machines, utilizing spare parking lots and old quarries, and perhaps the most desperate measure, emptying whatever was swept up by the plow's blade into the nearest waterway. Though expedient, given the visible dirtiness of urban snow even just a few hours after snowfall — plus the latent litter — this is hardly a desirable option from an environmental standpoint. From a cost one, though, it makes sense in certain cases, such as in locales with typically temperate winters. Where there is not much or any specialized snow-removal equipment, the closures necessitated have high costs. This means, depending on the forecast, simply waiting for the snow to melt (which can inspire strange conspiratorial hand-wringing when that doesn't work) is not always a sensible strategy — particularly when even the most-prepared of cities allows itself four or five days to fully remove 7" of new snow. And this with an annual budget of $145 million. Of course, even what's straightforward in principle becomes complicated once large numbers of people are thrown into the mix. Slow-moving trucks and improperly parked cars create obstacles to efficiency, while salt gets into waterways and onto floors. Expectations are high. This makes modern snow removal a complex operation, one that takes time to unfold. This time element makes the high snowfall rates of intense blizzards and lake-effect storms especially challenging; on the flip side, with preparation time being on the same order of magnitude as the storm length itself, by the time resources are mobilized and out patrolling the streets the forecast can have already changed — resulting in a general overestimation in years when eventual snow totals are low, as in the figure below. Coastal cities like New York tend to see large yearly snow totals due to big storms, rather than a series of light events, which pattern presumably explains the per-inch cost uptick for the snowiest winters. But snow removal presents challenges of a sort familiar to municipalities used to dealing with uncertain projections of population change, economic growth, political situations, and so on. Near-term weather forecasts are actually somewhat more accurate than economic ones (nicely analyzed in a reaction piece), though in large part this is because weather forecasters tend to hedge their bets much more than financial analysts do. Zooming out to a climate scale, the challenges of snow removal will tend to have less and less importance, as snow totals are expected to decrease in most cities around the globe, though with much regional variability due to storm-track shifts and the current winter temperatures in a region relative to 0 C. For example, northern Europe is expected to see more storms and higher wind speeds due to an intensified semipermanent low over Great Britain, but it is not yet clear whether the overall wintertime warming in that region will increase temperatures at the critical moments, i.e. the days immediately before and during intense storms, and consequently whether preparations for greater flooding or stronger snowstorms are more called-for in the British Isles and Scandinavia. In North America and possibly in Asia, mean snowfall is robustly expected to decrease, though with the caveat that localized areas (e.g. downwind of lakes) and events (i.e. the biggest storms) may depart significantly from this trend ["less common but more intense" is the phrase for what's often seen in models]. Taking a broader perspective, snow removal is a challenging aspect of municipal management in just a handful of major cities around the globe, and in generally wealthy ones at that, as seen in the neat time-lapse video from NASA's MODIS satellite below. And some cities would certainly be glad if snow removal were still the biggest thing their nations had to deploy the military against. Just about everyone professes to love media these days, and particularly its 'power'. But power can be destructive as well as constructive. The big question then is (at least for the purposes of this blog), does (mass-)media attention skew public perception of weather and climate, and of trends therein? Public opinion is important for its influence on policies from the international to local scales, and insofar as the quality of an opinion is proportional to the knowledge from whence it derives, public understanding matters too. A competent and engaged citizenry can particularly effect impacts in urban areas, whether the engagement is political, physical, or moral. In addition, broadly accessible, accurate climate knowledge is desirable from a philosophical standpoint, and a big part of determining that is the faithfulness with which scientific findings are filtered and transmitted onward to a wider audience. It's human nature, and consequently the nature of our media, to bring an issue sharply into focus for a while and then to have it recede into the distance of our collective consciousness. Events perceived as sufficiently worthy — because they are dramatic, heart-wrenching, concerning, etc. — draw our eyeballs and ears, then our hearts, and in some cases our time and pocketbooks. Then, barring new developments, some successor appears and displaces them. This sequence means that the public measure of a thing tends to rise and fall in exaggerated proportion to the oscillations of its 'true' measure — whether that thing is the price of a company's stock, the concern about a disease's impact (see figures below), or the state of the Earth's climate. There is considerable evidence that, when it comes to environmental issues, the intensity of media attention often does not match the deservedness of the story, but rather is determined by the political relationship between the country of the writers and the country being written about; the amount of drama that the issue presents; the wealth, race, etc. of the people affected; and so on. Although this general pattern is known from intuition, the degree of it is rather shocking: the death toll of a disaster in Africa must be 40 times that of one in eastern Europe to achieve the same level of worldwide coverage; multiply that factor by 1,000 and that's how much more the media focuses on fatalities from volcanoes versus droughts. That article also discusses how the prominence of stories rises and falls as the 'news cycle' goes around and as more-immediate issues crop up — for example, the percentage of Americans saying 'global warming is not happening' doubled from 2008 to 2010 as the economy fell into recession (compounded by "Climategate" in 2009). To the extent that the majority of the public's understanding of climate issues is framed by media coverage, there is significant potential for distortion relative to the scientific understanding. Films like the recent "Merchants of Doubt" are predicated on the assumption that people's grasp of the issues is indeed muddled by media portrayals (as opposed to, for example, representations of uncertainties about anthropogenic climate change and undue certainties about short-term forecasts merely reflecting independent public appraisals). There are also simply unintended consequences in perceived relevance of a topic when fluctuations in attention to it vary by an order of magnitude. A study analyzing word choice in articles about climate in U.S. and Spanish newspapers found the overall characterization of a weaker scientific consensus in the U.S. as opposed to Spain, matching public-opinion differentials, and further that over the 2001-07 period U.S. writers highlighted 'change' and 'surprise' with greater frequency (regardless of ideological bent) even as the scientific backing became more solid. It may be argued that this resulted from a desire to better characterize uncertainty, which is generally poorly understood; in any case, it was associated with a shift in U.S. public opinion in the opposite direction from that of scientists. Responsibility also partly rests with the scientists, of course, who as a group have not excelled in communicating to the public the complexities of their fields largely without the aid of explanatory figures and Q&A sessions that are the bread and butter of scientific conferences. A similar study in Malaysia published last month noted cyclical trends in climate coverage in local newspapers, in accordance with what is termed 'agenda-setting theory' — that media serve not just as actors in a broad societal pageant, but as directors who shape its contours. It's a tough job, as there are complicated and interlocking series of trade-offs: for example, focusing on anthropogenic impacts on the environment stimulates feelings of salience but paralyzes feelings of capability to address them. One of the main resulting discourses has been somewhat of a distraction: namely, who has the proper authority to speak as an expert. There is also substantial evidence from other fields that the balance of public opinion is swayed by media coverage independent of the true nature of the topic — partly a reflection of 'availability bias' (whatever has been heard the most becomes the dominant impression). Further contributing to confusion has been the recent apparent increase in media attributions of extreme weather to climate change that are only partially even possible — and qualifying this with remarks about its being 'unsophisticated', while well-intended, does not seem likely to materially affect the overall impression of a definite causation pathway in the mind of the layperson. A recent review article focusing on North America considered this sort of lack of nuance concerning, arguing that the urge to simplify paradoxically often leads to greater confusion, as many people consequently (and again this partly goes back to tenuous grasps of probability distributions as well as the power of visuals) fall under the impression that every local event is 100% attributable to climate change. This is true for long-term trends as well: an article in Climatic Change noted that two of the main complaints about climate change in Switzerland (low-snow winters and cloudy summers) are not actually supported by the data. Taken all together, this may help explain why media usage and environmental-problem action-taking are so weakly linked. Indeed, in some circumstances the media consciously work to define a framework characterized by a nationally motivated aversion to climate-change mitigation concomitant with an insistence on the existence of the problem so as to unify those who feel victimized, thereby using climate awareness merely as a means to another end — as in the case of India. Social media, in contrast to its mass counterpart, is more like personal experience in that it is less prone to being filtered through the subjective lens of 'worthiness'. In other words, in broadcasting the full range of reactions to a particular event, it is in a sense both more authentic and more biased. As discussed earlier, this sort of immediacy can be a positive as well as a detriment in terms of guiding public opinion toward the current numerical understanding of the climate, depending on the conditions at a particular place and time (especially those experienced by influential people). Local experiences are not to be underestimated when it comes to their power to shape opinion, at least temporarily. On the other hand, social media are prone to what has been termed 'siloization' or the 'echo-chamber effect'. I will write more on this in an upcoming post. As at very least an extension — and at most a fundamental constituent — of human culture, perhaps it is the prerogative of the media to engage in some amount of sensationalism in order to move public opinion and motivate public action to address the issues that seem most pressing, before they get 'out of hand'. Or, equally, to deem action undesirable. People in large part rely on the media of necessity as their source of information on many issues. Yet this all is an extraordinarily imperfect game for a multitude of reasons, including the natural focus on short-term risks, the various biases that determine what ends up receiving media focus, and the collective attention span of the public (which, it may be argued, is the driving force behind the pace of the news cycle in the first place). It may be that the system we have, and the discourse it engenders, is the best that may be reasonably hoped for given constraints like public scientific literacy and curiosity. An interesting question. At any rate, the pendulum of society-wide opinion is always swinging, and so reporting on the science with no agenda other than the aim for truth would hypothetically tend to move it back in line with scientific understanding. The only question is how much time that shift would require, assuming that it's even possible.
|
Archives
September 2023
Categories |