Confirming Global Climate Patterns by Exploring City Climates of Your Choice

We teach a 1000-level General Education course that introduces worldwide patterns of temperature, humidity and rainfall, and the 3-cell model of wind circulations. We want to create an assignment to accompany these modules where students can apply these theories governing worldwide atmospheric patterns to see how they impact the annual climate of a couple of specific cities. To make the assignment more engaging, we want the students to be able to choose basically any worldwide city. This also reduces issues of cheating and plagiarism.

The initial issue that came up is the difficulty of finding climate data for a wide range of worldwide locations. Our minimum requirement is the monthly average high and low temperatures and monthly average precipitation accumulation through a year. Having rain and snowfall separated would be a bonus. Ideally, the data can be presented in graphs with fixed axes so the climate of at least the cities the students choose could be easily compared. While some National Weather Service websites have nice charts, many are only for United States locations. I can’t find a National Center for Environmental Information website that presents basic running or monthly data for all months at one time. NCEI mainly shows comparisons between years for an individual month (e.g. July 2023 vs July 1999).

Without a more official source to go to, we found the website weatherspark.com, which uses models to create charts for worldwide cities. But their target audience is not students, and appears to be tourists. Their use of relative y-axes to fit the range of data and running 31-day precipitation totals make it difficult to come to meteorologically sound conclusions and to make comparisons between cities. Each graph is associated with automatically generated text, but their non-standard definitions and attempt at precision obscure the types of general conclusions we want students to arrive at. For example, according to the site Baltimore’s running 31-day rainfall totals range between 2.1 and 3.6 inches through the year. The graph appears to show consistent rainfall from April to November with a relative dry season between December and March. But the text says it has a 4.4 month wetter season from April 12 to August 24 and a 7.6 month drier season from August 24 to April 12. This text suggests a shift in late August that should last to December that isn’t reflected in the graph. Also, the text accompanying Baltimore’s snowfall totals report its snowy period lasts from November 27 to March 21, while its snowless period lasts the other 8.2 months. This precision seems to incentivize students to quote this text in their writeup, and hinders them from realizing the general principle that those 8 snowless months are summer.

Therefore, I attempted to address this problem by using the Global Historical Climatology Network (GHCN-D) worldwide station data to produce graphs of the most common meteorological variables in a way where their annual climate could be analyzed. Even better, I plotted the extreme ranges of data for all stations of similar latitude so a station’s climate could be compared to the global context. This is done for temperature, rainfall, snowfall, sunshine, and wind speed as long as enough data is available to represent a good climatology.

Specifically, grayscale color-filled areas indicate extreme values for four ranges of latitudes:

• The 5-degree band around the station. (e.g. station is at 45.3 N, band is for 43-47 N)
• The thirty-degree low, middle, and high latitude ranges. (e.g. 30-60 N)
• The entire hemisphere. (e.g. 0-90 N)
• All global stations. (90 S to 90 N)

Because the y-axes must span global extreme values, text is provided for each plot to indicate that station’s minimum and maximum values and what months those extremes occur in.