Chapter 4
The Humble Scale of Human Climate Influence
The earth's temperature results from a balance between warming by absorbed sunlight and cooling by infrared radiation emitted back into space. As the planet warms, it emits more infrared radiation until reaching "radiative equilibrium." The earth's reflectivity, or albedo, determines how much sunlight is absorbed versus reflected, with the planet currently reflecting about 30% of incoming sunlight.
Greenhouse gases like water vapor, carbon dioxide, and methane trap heat by intercepting infrared radiation from Earth's surface. Water vapor is the most significant contributor, but CO2 has increased from 280ppm in 1750 to 410ppm today. This seemingly tiny change (fewer than three molecules per 10,000) increases the atmosphere's heat-intercepting ability by about 1%.
When measured on the absolute Kelvin temperature scale, this 1% change in heat interception corresponds to approximately a 1% change in temperature (about 3C), showing the climate's sensitivity to small influences. Human activities also affect climate through other greenhouse gases, aerosols (which have a cooling effect), and land use changes.
Together, human influences amount to just over 2 W/m2 - about 1% of the natural energy flow through the climate system. This physically small influence requires extremely precise measurements to understand its effects amid natural variability.
The carbon cycle illustrates this challenge. Human-emitted CO2 represents a small addition to the vast natural carbon cycle moving among the earth's crust, oceans, plants, and atmosphere. The largest carbon reservoir is the earth's crust (1.9 billion gigatons), followed by oceans (40,000 Gt), land/soils/living things (2,100 Gt), and fossil fuels (5,000-10,000 Gt). The atmosphere contains about 850 Gt, mostly as CO2.
Human fossil fuel emissions contribute about 4.5% to the annual carbon flow, with about half absorbed by increased vegetation and half remaining in the atmosphere. This persistence is unique - about 60% of emissions remain in the atmosphere after 20 years, 30-55% after a century, and 15-30% after a millennium.
Chapter 5
Climate Models: Powerful but Problematic
Climate models are computer programs that perform mathematical simulations of the climate system. While powerful tools, they have significant limitations, as statistician George Box famously noted: "All models are wrong, but some are useful."
These models divide Earth into a three-dimensional grid - typically with 100km x 100km surface squares for the atmosphere (with 10-20 vertical layers) and 10km x 10km squares for oceans (with up to 30 layers) - creating about one million atmospheric grid boxes and one hundred million ocean grid boxes. Using fundamental physics, they calculate how air, water and energy move between boxes in small time steps.
However, many crucial phenomena like clouds and mountains occur on scales smaller than the typical 100 km grid size, requiring "subgrid assumptions" that vary widely between modelers. These assumptions introduce significant uncertainty, especially regarding clouds - the greatest source of modeling uncertainty.
Using finer grids would dramatically increase computation time; a simulation that takes two months with 100 km grid squares would take over a century with 10 km squares. Models must also be "tuned" by adjusting numerous parameters, a process that fifteen leading climate modelers described as "often seen as an unavoidable but dirty part of climate modeling, more engineering than science."
Climate assessment reports rely on ensembles of models that dramatically disagree with each other and observations. The models fail to reproduce the strong warming observed from 1910 to 1940, simulating only about half the actual warming rate. The IPCC admits "we've no idea what causes this failure" - deeply unsettling since this early warming is comparable to the late twentieth-century warming confidently attributed to human influences.
Natural climate cycles like the El Nino-Southern Oscillation, Atlantic Multidecadal Oscillation (AMO), and Pacific Decadal Oscillation (PDO) create internal variability that makes it difficult to determine which observed changes are human-caused versus natural. While models can reproduce some aspects of El Nino events, they struggle with slower cycles.
Chapter 6
Extreme Weather: The Gap Between Headlines and Data
The media and politicians routinely blame extreme weather on climate change, but the scientific evidence presents a more nuanced picture. Most extreme weather events show no significant trends attributable to human influences, with multiple IPCC reports expressing "low confidence" in global trends for floods, droughts, severe storms, and extreme cyclones. This disconnect between public perception and scientific data highlights the importance of examining actual meteorological records rather than relying on news coverage.
Temperature records reveal particularly interesting patterns. Despite popular perception, record daily high temperatures in the United States are no more common than a century ago, though record lows have decreased significantly. Professor John Christy's comprehensive analysis of absolute record temperatures from 725 US weather stations since 1895 confirms this pattern. His research shows that record highs demonstrate no significant trend over 120 years, while record cold temperatures have been declining steadily. For example, in the 1930s, states like Kansas, Oklahoma, and Missouri experienced some of their highest temperature records that still stand today. The evidence indicates that temperature extremes in the US have actually become less common and somewhat milder since the late nineteenth century, characterized by fewer harsh winters and cold evenings rather than more torrid summers.
Hurricane patterns present another example where data contradicts popular narratives. Despite theoretical expectations that warming seas would increase hurricane activity, the historical record shows complex patterns with no clear human influence. North Atlantic hurricane activity follows natural cycles, particularly the Atlantic Multidecadal Oscillation (AMO), which produces alternating periods of high and low activity lasting 20-30 years. For instance, the active hurricane periods of the 1940s-1960s mirror recent patterns, suggesting natural variability rather than anthropogenic change. A landmark 2019 paper authored by eleven leading tropical cyclone experts found minimal evidence of detectable human-caused changes in hurricane patterns, despite analyzing over a century of data.
Tornado statistics tell an equally compelling story about the gap between perception and reality. When examining only EF1 or stronger tornadoes (those causing significant destruction), there's no trend over sixty years of reliable records. More tellingly, the strongest tornadoes (EF3+) have actually decreased by approximately 40% since 1954. States in "Tornado Alley" like Oklahoma and Kansas have seen no increase in severe tornado frequency. The most significant tornado-related change has been unambiguously positive - annual US tornado deaths have fallen dramatically, from about 80 people per million in 1875 to fewer than 8 per million today, thanks to improved radar systems and warning networks. Modern Doppler radar technology now provides average warning times of 13 minutes, compared to virtually no warning in the early 20th century.
This careful examination of long-term weather records demonstrates the importance of distinguishing between media coverage and scientific evidence when assessing climate-related changes in extreme weather patterns.
Chapter 7
Precipitation Patterns and Sea Level Rise: Context Matters
Global precipitation data shows minimal change - only about 0.2% increase per decade since 1901, with high variability making this trend statistically weak. US precipitation has increased slightly (0.6% per decade) since 1900, but with dramatic regional variations. The year-to-year fluctuations far exceed the modest overall trend, making it difficult to distinguish human influence from natural variability.
Despite modest rainfall changes, US flooding trends vary geographically. Globally, the IPCC expresses "low confidence regarding the sign of trend" in flood magnitude or frequency - essentially admitting we don't know whether floods are increasing, decreasing, or unchanged worldwide. Historical records show floods larger than modern ones occurred throughout Europe, the Mediterranean, and East Asia during the past 500 years.
Drought assessment is even more complex. The Palmer Drought Severity Index for the contiguous US shows no significant long-term trend since 1895, with the past fifty years actually slightly wetter than average. Tree-ring data reveals the Southwest experienced multiple decade-long "megadroughts" between 900-1300 AD during naturally warmer periods, far exceeding anything in the 20th century.
Sea level rise presents a similar story of context and perspective. Despite alarming media imagery, actual tide gauge records at Manhattan's Battery show sea levels rising at only about 30 cm (1 foot) per century since 1855. While recent sea level rise rates are higher than the 20th century average, they must be viewed in context of substantial historical variability. The IPCC itself acknowledged that similar high rates occurred between 1920-1950.
Local sea level is what truly matters for planning adaptation measures, and these differ from global averages due to factors like land subsidence, tectonic activity, and ocean currents. Future projections remain highly uncertain due to limited understanding of ice sheet dynamics, with the World Climate Research Programme acknowledging "major gaps in our understanding."
Chapter 8
Apocalyptic Claims vs. Scientific Reality
The media and political discourse attribute numerous impending catastrophes to human climate influence - from death and destruction to disease, agricultural collapse, and economic ruin. However, historical data doesn't support such claims, and projections of future impacts stem from implausibly extreme scenarios fed into unreliable models.
Consider climate-related deaths: In 2019 testimony to Congress, economist Michael Greenstone claimed climate change could cause an additional 85 deaths per 100,000 by 2100. This alarming projection represents six million additional annual deaths globally. However, examining actual data reveals weather-related death rates have fallen dramatically during the past century even as global temperatures rose 1.2C - they're about 80 times less frequent today than a century ago.
Similarly, despite alarming headlines about climate change threatening food supplies, global production of both crop and animal calories has increased dramatically since 1960, with yields of wheat, rice, and maize more than doubling in fifty years. Rising CO2 concentrations have actually boosted crop yields through enhanced photosynthesis and water efficiency, while "greening" 25-50% of vegetated areas globally.
Economic impacts tell the same story. The UN's Fifth Assessment Report shows that a global temperature rise of up to 3C by 2100 would negatively impact the global economy by only 3 percent or less - translating to a minuscule 0.04 percent annual reduction in growth rate. Even the 2018 National Climate Assessment projects that a very large warming of 5C would diminish the US economy by just 4 percent by 2090. Put in context: assuming conservative 2% annual growth, the US economy will quadruple by 2090, and this "devastating" climate impact would merely delay that growth by about two years.
Chapter 9
Why Is Climate Science So Poorly Communicated?
If crucial parts of climate science remain unsettled, why is the narrative of "The Science" so different from actual science? The disconnect stems not from some secret conspiracy but from a self-reinforcing alignment of perspectives and interests among various stakeholders in climate matters - scientists, institutions, activists, media, and politicians. This complex web of interests creates an echo chamber where certain viewpoints are amplified while others are diminished.
Media coverage of climate science has become increasingly provocative to drive clicks, with climate coverage particularly susceptible to sensationalism. News organizations face business pressures that favor alarming stories over nuanced reporting on scientific realities. For example, headlines about "climate catastrophe" generate more engagement than articles discussing uncertainty ranges in temperature projections. Staffing changes have reduced in-depth science coverage, with many newspapers eliminating dedicated science desks in favor of general assignment reporters. The emergence of dedicated "climate reporters" often creates a predetermined narrative of doom, as these journalists frequently lack scientific training and may approach stories with advocacy rather than analytical mindsets.
Politicians exploit climate issues to arouse voter passion, presenting complex scientific matters as simple crises requiring immediate action. Climate threats are politically useful because they resonate universally while allowing predictions that won't be tested until long after politicians leave office. This dynamic encourages bold claims about future impacts without accountability for accuracy. Politicians can propose sweeping policy changes while citing "consensus science," even when the underlying research contains significant uncertainties or caveats.
Scientific institutions often prioritize narrative over accuracy in climate communications, perhaps fearing that acknowledging uncertainty might undermine public support for climate action. Despite the National Academies' reputation for objective, high-quality advice, their climate assessment reviews have frequently fallen short of their usual rigorous standards. Such statements bypass normal review procedures while carrying institutional weight, undermining public trust in scientific pronouncements. The result is a growing credibility gap between institutional pronouncements and the actual state of climate science.
Individual climate scientists face institutional pressures that discourage honest communication about uncertainties. Career advancement, funding, and peer acceptance all push researchers toward producing alarming findings that generate press coverage. Grant proposals emphasizing potentially catastrophic outcomes are more likely to receive funding than those suggesting moderate impacts. As MIT oceanographer Carl Wunsch noted, scientists analyzing inadequate climate data face difficult choices: admit the limitations and risk losing grants, or distort uncertainty calculations to produce headline-grabbing conclusions. This creates a selection bias favoring more extreme predictions and interpretations of data.
The peer review process itself has become politicized in climate science, with researchers reporting pressure to remove or soften caveats about uncertainty. Conference presentations and journal submissions that challenge prevailing narratives face heightened scrutiny, while those supporting consensus views often receive less rigorous review. This institutional environment makes it professionally risky for scientists to publicly discuss uncertainties or limitations in climate science, creating a false impression of greater scientific certainty than actually exists.
Chapter 10
Moving Forward: What We Could, Should, and Will Do
Given the enormous challenges of effectively reducing emissions and the concerns surrounding geoengineering, adaptation to climate change will likely complement, if not overshadow, mitigation efforts. This isn't a statement about what we should do, but rather a judgment about what we will do.
Adaptation offers several advantages: it's agnostic (works regardless of climate change causes), proportional (can be scaled as needed), local (tailored to specific needs without requiring global consensus), autonomous (happens naturally as societies evolve), and effective (societies already thrive in diverse climates).
Koonin suggests several concrete steps forward:
1. We need sustained, improved climate observations across all systems to understand climate changes that occur subtly over decades.
2. Climate models require better understanding rather than more unproductive computing - we should focus on why they fail to describe the recent past and make uncertain future projections.
3. The science needs improvement through open, honest discussion beyond slogans and polemics, possibly through a "Red Team" exercise scrutinizing upcoming reports.
4. We should pursue "easy" emissions reductions like stopping methane leaks, reducing exotic greenhouse gases, implementing cost-effective efficiencies, and researching emissions-lite technologies.
5. We should pursue adaptation strategies more vigorously, as humans have adapted to diverse climates throughout history.
6. We should promote economic development in developing countries to improve their adaptation capacity.
7. We should research geoengineering options as a prudent precaution against significant climate deterioration.
What we should do, in short, is restore integrity to climate science and take steps most likely to result in positive outcomes for society. The socio-technical obstacles to reducing CO2 emissions make it likely that human influences on the climate will not be stabilized this century. This isn't "waffling" but "realistic" and "prudent" - we should make only low-risk changes until we better understand climate change.
