In examining historical climate data, particularly from millions of years ago, we find indications that global warming might escalate beyond what current models forecast. The reliability of these "proxy" models is under scrutiny.

There's no denying that the planet is heating up. Climate models illuminate the reasons behind this phenomenon. The combustion of fossil fuels releases carbon dioxide (CO2) that has been trapped underground for eons. Despite constituting only a small portion of the atmosphere, CO2 acts like a toxin for our climate, with increases of 100-200 parts per million (ppm) potentially leading to disastrous outcomes.

To gauge the severity of these impacts, we rely on models that estimate how much warming will occur with increased CO2 levels. While this is a complex task, current models have successfully predicted temperature rises observed over the past three decades.

However, warming is not an isolated event; it influences ocean currents and contributes to the melting of sea ice. Over time, human activities have also modified the climate in ways that are not solely connected to CO2 emissions.

Recently, we've witnessed failures in climate models. For instance, the year 2023 was significantly warmer than predictions—by approximately 0.2 degrees Celsius, a considerable difference when considering the entire planet, including oceans. This discrepancy has cast doubt on climate models that had remained consistent for four decades.

While climate change denial is akin to ignoring evidence, questioning the accuracy of yearly temperature predictions is a valid concern.

The complexity of climate science extends beyond CO2 alone. If it were that simple, understanding climate change would be much more straightforward. There are numerous factors that can also influence climate patterns, some of which we can incorporate into models, while others elude our grasp.

One theory for the unexpected warmth in 2023 is linked to a reduction in atmospheric aerosols, such as sulfur dioxide (SO2), particularly from global shipping. These aerosols, which also come from volcanic eruptions and wildfires, have historically had a cooling effect on the climate. As we've reduced these pollutants without similarly decreasing CO2 emissions, we may have inadvertently removed this cooling influence.

A recent Washington Post article discussed this phenomenon in depth.

Despite acknowledging the decline in aerosols, models still struggled to explain the extreme heat of 2023. NASA identified four key factors that contributed to last year's warming, ranked by their significance:

1. Rising CO2 levels, unsurprisingly.
2. The influence of El Niño and La Niña events. These phenomena dictate yearly climate variations, with La Niña years typically being cooler and El Niño years causing warmer temperatures. The transition from a prolonged La Niña to a strong El Niño in 2023 led to unprecedented warming.
3. Long-term trends of ocean warming. The oceans have absorbed roughly 90% of the heat from global warming.
4. Aerosols, which primarily affect specific regions rather than the globe as a whole.

The critical question arises: why haven't models accounted for these variables?

Climate models have struggled to predict the intensity and duration of El Niño and La Niña cycles until they are already in progress. They failed to foresee three consecutive years of La Niña and, although the subsequent El Niño was weaker than those in 1997 and 2015, it still ranked among the strongest recorded.

Without this information, the models were unable to accurately forecast global temperature increases. Artificial intelligence shows promise in improving these predictions, potentially allowing for more accurate year-to-year climate assessments.

Nevertheless, these fluctuations do not alter the overarching trend of climate change. Using a single anomalous year to invalidate the entire modeling process lacks intellectual integrity.

To better understand future climate scenarios, we should look to paleo-modeling—analyzing ancient climates and their implications for the present.

Much of our understanding of historical climates remains elusive. We lack reliable CO2 estimates prior to approximately 800,000 years ago, as this is when we first have ice core samples. Before this timeframe, we must rely on geological records and make inferences about atmospheric conditions.

In the past 800,000 years, CO2 levels have never approached those of today. However, periods exist in the distant past when CO2 concentrations were significantly higher, resulting in average temperatures so elevated that even the Arctic supported alligators.

Take, for instance, the Eocene epoch about 55 million years ago. This period experienced extreme warmth, with temperatures approximately 8 degrees Celsius higher than today. In stark contrast, even a 2-degree rise in temperature today would lead to catastrophic consequences.

During the Eocene, CO2 levels were four to five times greater than present levels, coinciding with intense volcanic activity. The planet had no polar ice caps, and sea levels soared by 70 to 140 meters (230 to 560 feet).

This historical context provides insight into Earth's climate history, which oscillates between greenhouse climates (with no polar ice) and icehouse climates (which do have ice). Currently, we inhabit an icehouse climate phase.

Under greenhouse conditions, tropical and subtropical flora and fauna could thrive at high latitudes—think Greenland, not New York. Conversely, icehouse climates do not support such biodiversity.

Fossil evidence from the Eocene suggests that primate species akin to today's lemurs once inhabited the Arctic, specifically on Ellesmere Island:

This region now has average July temperatures of merely 6.4 degrees Celsius (43.5 degrees Fahrenheit), and the winters are even harsher.

A significant aspect of global warming is the uneven distribution of temperature increases across the planet. Higher latitudes experience warming at a faster rate than lower ones, as heat moves from hotter equatorial zones to cooler polar regions. An excellent example of this is the Gulf Stream, which carries warmth from the Gulf of Mexico to Europe, making it milder than similar latitudes in North America and Asia.

As CO2 levels decreased following the Eocene, Earth cooled, although it remained warmer than today. By the onset of the Oligocene, around 33 million years ago, polar ice sheets began to form in Antarctica, nearing their current size. By the middle of the Oligocene, CO2 concentrations had settled into relatively lower levels, ranging from 300 to 700 ppm.

This should have signaled the beginning of an icehouse climate, and technically it did, but the situation became perplexing. The climate remained warm despite the presence of ice caps. Deep-sea drilling samples indicated no significant transition in sea surface temperatures.

As CO2 levels neared modern levels around 26.5 to 24 million years ago, instead of cooling, the climate warmed by 1 to 2 degrees Celsius.

It's crucial to recognize that throughout much of Earth's history, CO2 levels serve as a reliable indicator of global temperatures. During the late Oligocene, sea surface temperatures in tropical regions mirrored those of the exceedingly warm Eocene era, while polar temperatures were unexpectedly high as well. The temperature differential between the equator and poles was much narrower than it is today.

Even models designed for the Eocene frequently fail to accurately represent the Oligocene era. Regardless of the assumptions made, they consistently underestimate temperatures at high latitudes. The reasons behind the sustained warmth of the Oligocene climate remain a mystery.

Recent estimates suggest that Oligocene global mean surface temperatures ranged from 22 to 24 degrees Celsius, while the Eocene averaged around 23 degrees Celsius—8 degrees higher than the current average of 15 degrees Celsius. Conversely, deep ocean temperatures remained low, indicating that the heat source was not from the ocean.

This creates a paradox: we have contradictory data regarding the Oligocene climate that conflicts with (1) cool deep ocean temperatures, (2) the existence of polar ice caps, and (3) climate models that fail to explain these anomalies.

Currently, the reason behind the unusual climate behavior during the Oligocene eludes us. Factors such as ocean circulation patterns and cloud feedback may play a role.

For today's climate context, the implications are clear: as CO2 levels rise and contribute to global warming, we cannot wholly depend on ancient climate models to predict future outcomes. The late Oligocene's CO2 levels were similar to those of pre-industrial times, yet average temperatures may have been 7 to 9 degrees higher. Such an increase would have devastating effects on modern human civilization and countless animal species adapted to our current, relatively cooler climate.

It would be naive to assume that proxy models from the past are directly applicable to our present circumstances. The fact that similar CO2 increases in the past resulted in significant temperature rises millions of years ago does not guarantee the same effects today, especially as we lack a comprehensive understanding of all the contributing factors. We must prioritize modern climate models over ancient proxies, as there remains much about our climate—past, present, and future—that is still unknown.