Overnight lows on June 15, 2022
Climate Central’s Realtime Fingerprints program develops tools like the Climate Shift Index (TM) to help reveal how climate change is disrupting our world.
Our first operational tool, the Climate Shift Index (TM) provides daily estimates of how climate change is impacting air temperature. It builds on a decade of research on climate change attribution to put a number on the influence of climate change on daily high and low temperatures.
Frequently Asked Questions
The Climate Shift Index ™ is a number that indicates the fingerprint of climate change on any day’s local weather. The initial implementation of CSI is for high and low air temperatures. A CSI level above zero means that human-caused climate change has made that day’s temperature more common (and a level below zero, less common). A CSI can be calculated for observed temperatures and for forecasted temperatures.
Not quite. It’s a shift in how frequently certain temperatures occur. For example, suppose 90°F used to occur on average one day each year in June where you live. If climate change has altered the conditions so that 90°F now occurs on average three days each June, then its frequency (or likelihood) has increased by 3 times. Following this estimate, whenever we see a 90°F day in June in your location, it is assigned a CSI of 3.
The CSI is a categorical scale, with the categories defined by the ratio of how common (or likely) a temperature is in today’s altered climate vs. how common it would be in a climate without human-caused climate change. For the positive CSI conditions (which occur much more often than the negative), we assigned a simple descriptor to these events (see table).
We built our system to work with temperatures that are observed most often. Level 5 events on our scale are exceptional events that are candidates for more detailed analysis, including the special statistical techniques necessary to work with extremely warm (or cold) temperatures beyond what has been observed. In 2021, 99% of the daily high temperatures in the U. S. had a CSI from -4 to 4.
To compute the CSI, we combine two complementary ways of estimating how climate change has altered the frequency or likelihood of a temperature. We average these two methods together to make the CSI. One method uses 22 climate models each run with and without historical greenhouse gas emissions. We then calculate the frequency of the observed or forecasted temperature in the modeled climates. This gives us 22 pairs of frequencies. We then take the ratio of the frequency in the high-CO2 run to the frequency in the run without extra greenhouse gasses. We then average these ratios together.
For the other method, we take all of the daily temperatures for the last 30 years and use that to calculate the frequency of the observed or forecasted temperature. We then calculate how much a particular location’s temperatures change in response to an increase in global mean temperature. This relationship allows us to remove the contribution of human-caused climate change, giving us the frequency of the particular temperature in a climate without global warming.. We then compute the ratio of the frequency of the temperature in the current climate to the frequency in the climate without climate change. We have two different ways to estimate the current and past frequencies. These two values are averaged together. The final Climate Shift Index is built from the average of the observation-based and model-based estimates.
See our downloadable methods documentation for more detail.
Not entirely. Any weather event has multiple causes. The CSI tells us how much climate change has boosted the odds of a particular temperature. Events where the CSI reaches level 5 would be very difficult to encounter in a world without climate change–-not necessarily impossible, just very, very unlikely.
The goal of our system is not to recreate the climate of 1850 or some earlier period. Instead, the goal of our system is to calculate how human-caused climate change has altered the climate. There are lots of factors that can affect climate in a location. For example, turning a grassy field into a city can make higher temperatures more likely, while allowing trees to grow back will make warm conditions less likely. We want to include these effects but remove the effect of greenhouse gasses. We do this by characterizing today’s climate using the best available data and then using the model- and observation-based techniques described above to remove the effect of the extra greenhouse gasses released by humans.
Our main goal when creating the CSI was to distinguish temperatures that have been influenced by climate change from conditions where that link is not firmly established. To make sure this is the case, if either of the two observation-based methods or the average of the model-based methods disagree on whether climate change played a role, then that event receives a CSI level of 0. This means that for any event with a positive CSI, there is strong evidence that climate change has contributed to those temperatures.
When talking about warming, it is natural to focus on the daytime high temperatures. However, climate change is also occurring at night and we see some of the strongest climate signals in these low temperatures. Elevated nighttime temperatures, especially during the summer, can create uncomfortable conditions and limit opportunities for people to cool down. Nighttime heat poses a health risk, especially for vulnerable populations such as the sick, elderly, and lower-income communities who may not have air conditioning. For people with air conditioning, elevated nighttime temperatures increases their utility bills.
While climate change is making hot days more likely, cold days still occur. The CSI tells you that climate change is making these cooler days less likely. The CSI is defined so that a temperature event with CSI level of -2 means that it is two times less likely (equivalently, they occur half as often) due to climate change.
Climate Shift Index ExamplesExplore the Climate Shift Index
High temperatures on June 17, 2022