A new study published in the Journal of the Association of Environmental and Resource Economists warns that errors in temperature measurements used in climate research might be causing scientists to underestimate the impacts of climate change by as much as 15%. The research was led by Alex Yu, an assistant professor of economics at Wake Forest University, and Richard Carson, a distinguished professor of economics at UC San Diego. The study highlights that the methods researchers use to estimate local temperatures—such as in cities, neighborhoods, or farmland—can introduce significant errors that vary depending on geography and location. For example, in mountainous areas like Boulder, Colorado, standard weather measurements differed from actual temperatures by nearly 15 degrees Fahrenheit on average. In contrast, in flatter and more densely monitored areas like Chicago, the temperature errors were as low as 1 degree Fahrenheit. These discrepancies can remain consistent over time and are influenced by local geography. When these errors are applied to real-world outcomes, such as crime rates or the number of heat-related illnesses, they can lead to misleading conclusions. The researchers used over 20 million crime records to show that the estimated impact of extreme heat on violent crime appeared smaller when measured at a broader geographic scale. This suggests that if climate studies underestimate the effects of extreme heat on crime, crop loss, or health risks, policymakers might not fully grasp the scale of interventions needed to address these issues. The study offers practical advice for researchers, noting that no single method for measuring temperature is universally effective. In areas with many weather stations, local measurements can be more accurate than complex computer-generated maps of temperature. In regions with fewer monitoring stations, these maps may be more useful. The research underscores the need for better spatial temperature monitoring, including expanding and maintaining dense weather station networks. By ensuring that temperature measurements align with where people, farms, businesses, and ecosystems are actually located, the study suggests that measurement errors can be reduced from the start.