FAQ

Frequently Asked Questions

I think you’re making this up. Where does your data come from?

All data comes from https://www.ncei.noaa.gov/cdo-web/ (Climate Data Online) and then searched for Daily Summaries. The other options (Global Summary of the Month, Normals Daily, etc) didn’t interest me because I wanted to analyze the data for myself and not accept someone else’s conclusions.

I also chose to limit my data searches to widely available data sets in consistent data formats instead of trying to get data from each individual city or newspaper or local historian. Anyone who wants to dispute my numbers is welcome to download from this source and compare my results.

I think that they are part of a conspiracy to prove climate change is really happening so how can you trust their numbers?

The data collected for this website, approximately 450,000 data points, is a tiny fraction of total amount of data. Because most cities did NOT have a single continuous data set reaching from the start to the current ate, more than 150 individual data sets were downloaded out of an estimated 10,000 data sets, which puts the total number of individual data points in the hundreds of millions.

The absolute scale of falsifying this volume of data is monumental. While computers could be programmed to simply add 1 degree to every temperature, this wouldn’t show any trend towards warmer overall temperatures. Because the temperature change trends are not consistently shown in every city, it would take a huge and extremely sophisticated computer program to manipulate the data in the subtle and inconsistent manner that exists in the data, a large enough effort that it seems unlikely that it could have escaped notice by the general public.

I think you cherry-picked the data . . .

There’s some truth in that. Some cities have 300+ data (Boston: 376) sets to choose from, others have just a few. When selecting data sets to download, I used the following criteria:

• As old as possible
• As long a span as possible (as many continuous years)
• As closely physically located as possible when a multiple data sets were required (many data sets could be separated by 50 miles or more which could lead to misleading data sets if, for example, one data set came from the beach town of Santa Monica and another data set came from downtown LA and another data set came from mountains near Big Bear.)
• As good quality as possible. For example, some data sets seemed to have data reaching back into the 1800’s, but when they were downloaded turned out to have no data until 1901. Others had missing months or even years. One data set had high temperatures of 333 degrees on the last day of the month for four months in a row (I used the average temp of the day before and the day after). Another data set had a temperature of -89 in the middle of summer (they had accidentally put in a negative value – removing the minus sign made the temp fit right in).
• When multiple data sets were spliced together to build a continuous history, an overlap of one to six months was included in the data to “smooth” the splice. During the overlap, both temperatures were used so the average temp appears in the charts.

Why did you limit the data sets to just the downtown areas?

The data sets selected were as close to the downtown area of each city as could be located based on the criteria described above. Based on their locations, many good candidate data sets were excluded because they only covered 5 or 10 years. One suspects that city budgets may interrupt temperature collection.

There is also a limit to the time or effort that I can devote to this study. Not to mention the difficulty of trying to present the results of these studies in a meaningful way, one that doesn’t bog the user down in details.

Aren’t you worried about the ‘heat island’ effect?

Just because urbanization may have influenced temperatures doesn’t mean that the increased temperatures didn’t happen. The term ‘heat island’ is inadequate because any increase in temperatures ‘flows’ towards cooler areas and don’t respect any city boundaries. One suspects that the extra heat absorbed by blacktop streets and concrete sidewalks during daylight hours and then released after sunset flows to surrounding areas, warming their nights and ‘preloading’ these areas for warming daytime temperatures.

The ‘heat island’ effect is imagined to be similar to a campfire: hot when you’re right next to it but can’t be felt a few yards away. In the 1800’s that certainly would have been an accurate analogy. However, in the 21st century, with the spread of (sub)urbanization, that campfire is more like a forest fire spread across thousands of square miles and sharing that additional warmth with surrounding areas. Worse than a forest fire that consumes the fuel in one area before moving on, (sub)urbanization collects heat every day and shares that extra heat every evening.

Aren’t you worried that your study is too limited?

It is definitely limited and certainly too small to make meaningful predictions, but it is useful as an indicator, as a way of understanding the current situation, as a way of fact checking the climate authorities.

Some areas to investigate in the future are:

• Compare areas before and after urbanization arrived and try to see if there are detectable changing in their temperatures
• Examine rural records and look for trends towards increased temperatures.
• Examine European records to see if their more compact cities are experiencing the same temperature trends as their rural areas.
• Compare oceanic events, such as El Nino’s, with land temperatures to try to identify correlations.
• Study precipitation (rainfall, snowfall, etc) in addition to temperature. For example, when did the first and last snowfall happen over time?