Weather Methodology

Updated 2016.05.10

On This Page:



These numbers aren't official in any way and aren't intended to contradict any numbers you see on local news broadcasts or any weather websites — none of those temperatures were recorded at my house.


All temperatures seen here are from my own thermometer and rainfall totals are from my own gauge, at my own house, recorded in South Central Killeen (except where noted). Rainfall totals are from my own rain gauge. Wind speeds and humidity are from ¹ and ⁴.

When my own gauges are unavailable (for example: on vacation, or before I had them), I used the following alternate sources, cited with superscript notations:

When ³ is used, it means I got numbers from one of the other sources and corrected based on their usual difference from my own numbers. For example, high temperatures from are often 3F different from my own readings, usually lower than my own. Based on a sampling of known temperatures, I've estimated that high temperatures are — on average — 2F warmer at my house than at the station. Lows deviate more widely, averaging three degrees or more warmer at my house.’s numbers are closer to my own but still fall into a known deviation pattern.

From 2013 onward, better than 98% of temperatures and rainfall totals are from my own gauges. Prior to 2013, I only recorded the temperatures occasionally, usually if it was very hot or very cold, often only recording the high for a hot day or only the low for a cold day.


On all my weather pages, temperatures are presented in degrees Fahrenheit (F), wind speeds are miles per hour (mph), precipitation is in inches (), humidity is percent (%).

In 2013, I began listing only the average humidity, but then switched to low-ave-high for a while. In late 2014, I switched to just low-high. In December 2014, I ceased listing wind speeds and humidity, realizing I was only [mainly] interested in temperature and rainfall trends.


Abbreviations used on my weather pages include the following:

  • CS = clear skies
  • MS = mostly sunny
  • PC = partly cloudy
  • MC = mostly cloudy
  • OC = overcast
  • FG = foggy (added 2015.11.02)
  • HLA = high/low average (average of the high and low temperatures)
  • +/- = change from HLA of same day/month of previous year

I started calculating daily HLAs in 2014, wanting a better way to compare one day (or month) to another day (or month). For example, July 15 of one year might have a high of 100F, and July 15 of another year had the same high. But one of them had a low that was ten degrees cooler than the other. Comparing only the highs, or only the noon temperature, or only the midnight temperature, would be misleading. One day was clearly cooler than the other. Using HLAs, I can say one of them had an HLA of 90F and the other’s HLA was 85F. Obviously, it would be even more accurate to record and average the temperature every hour or even every minute, but I don’t have the ability (or time) for that. The HLA method is my best compromise between complete accuracy and ease of use.

Not long after, I began going back to calculate the daily HLAs for 2013 and other years, and then doing whole-month HLAs. This was (again) to get more accurate comparisons. For example, July 2011 (HLA: 90:48) was hotter than July 2013 (HLA: 85.40).

Color Coding

I use colors for various indications on my weather chart pages. In general, the superlatives (coolest, warmest, etc.) of a particular column are indicated by bold typeface on a black background, with a color (blue, green, red). The same with a white font is a median for that column. A color used without bold and without black background, indicates one of two things: (1) second place for a column — second warmest, for example — or (2) upward/downward trend for the HLA (+/-).

above ave.2nd leastwarmerwarmestwarmer
below ave.2nd mostcoolercoolestcooler

Below is a screenshot of part of October 2016, with annotations in yellow, explaining briefly what each part means.

The image above represents how I was doing everything as of October 2016. In earlier years, the columns would have looked different.


In all cases, precipitation was measured just after the rain, instead of waiting until some predetermined time. In the past, I've watched half an inch of water completely evaporate from my rain gauge, so I've made sure to measure and record the amount as soon as the rain quits falling.

I’ve been told that some weather-watchers are instructed to check the rain gauge once a day, at predtermined times, such as 7 a.m. If I did this, my rainfall totals would be almost zero for most months — and they’re already lower than official totals in most cases.

Record high / Record low

When a daily temperature is marked as a "record high" or "record low", that is according to any sources I could find. This usually means ¹ and ², since no other sources of temperature records are available online (as far as I know). The National Weather Service does not make its daily high/low records available for Killeen (or any other city). Local TV stations are not based in Killeen, so when they give a daily record on the evening broadcast, it’s for Waco, or Temple, or Belton. I have contacted two of these stations, asking whether they have historical weather data for Killeen; neither responded.

Also, occasionally, the record was later superceded, but I've left the old notation in place.

I’m certain the records on my temperature records page are not all correct, because some of do no seem high or low enough to hold the record, but they were listed as records on the sites that make information available.


Note, for historical purposes, my house’s location (31.047440, -97.722606) was not in the city limits of Killeen until [date unknown]. When the city was smaller, this location was south of the city limits.

Reason For Interest

I’ve been interested in weather since childhood (1970s) when I noted that some winters had snow and others didn’t, and some summers were hot and dry while others were warm and wet. I began noticing patterns and wanted to make sure my observations weren’t subjective. One pattern I noticed since moving to Central Texas was the tendency to have a record-breaking hot year followed by two or more years of gradual cooling, and then another record-breaking hot year. Such a pattern gives the impression to residents that the climate overall is cooling, interrupted occasionally by very hot years, when in fact the opposite might be occurring — each very hot year is hotter than the last, interrupted by the deceiving cooling years. The only way to be sure was to start keeping track, daily.

I did not want to start keeping track daily. What I would have much preferred is for other weather stations to make their historical data available online, but most don’t. Wunderground makes it available, but there are missing days and days when their numbers are clearly inaccurate (one day showed 1300 inches of rain, for example, while another day showed the high as 81F when the entire area was known to have been above 100F). Plus, they don’t go back very far. Local TV stations, newspapers, government weather stations, etc. — none of them (that I’ve yet found) make available historical daily weather information.

For example, several sites will say the average annual precipitation for Killeen is about 33 inches, but none of them list rainfall totals per year. And the “average” hasn’t gone down, despite many consecutive years of below-average rainfall.

comments powered by Disqus