Common weather vs averages (hot, records, locations, season)

[alphatier]

Nice charts here. I wish I had better excel skills.

My average high for 2002 - 2015 for July is around 26.2C.

During this period 231 days are >= 26C and around 200 days have temperatures less than the average. 30 days are below 20C and 116 days >= 30C.

I'd like to make a chart but I have really no clue how to use this program.

[nei]

Originally Posted by ben86

Do you get similar departures from average for record highs/lows in summer?

Here are our record highs/low highs for the summer months vs the averages:

June:
Average high: 16.9C
Record high: 29C (rounded) - 12.1C above
Record low high: 7C (rounded) - 9.9C below

July:
Average high: 19.1C
Record high: 31.0C - 11.9C above
Record low high: 10.7C - 8.4C below

August:
Average high: 18.7C
Record high: 31.6C - 12.9C above
Record low high: 10.3C - 8.4C below

============================

You can see from the red graph, that it's symmetrical and with one peak, suggesting the mean and median are close.

Hmm. But checking:

June:
Average high: 78.2°F
Record high: 101°F
Average yearly extreme high: 91°F
Record low high: 46°F
Average yearly extreme low high: 61°F

July:
Average high: 83.0F
Record high: 104°F
Average yearly extreme high: 93°F
Record low high: 59°F
Average yearly extreme low high: 70°F

August:
Average high: 81.0°F
Record high: 100°F
Average yearly extreme: 91°F
Record low high: 56°F
Average yearly extreme low high: 67°F

I added average yearly extreme as the sample size is greater. But there's a slight skew in the opposite direction — slightly more extreme colder than average temperatures than warmer. But outside of extremes, the distribution is symmetrical and mean = median. Those days with cool highs are typically from very cloudy times when it rains all day.

[nei]

Originally Posted by alphatier

Nice charts here. I wish I had better excel skills.

I didn't use Excel for my charts.

[alphatier]

What do you use?

[nei]

MATLAB for those charts, I switch back and forth between the two. Computer code is more efficient for large datasets, especially if doing similar graphs and processing. R is a good free alternative.

[Donar]

I used Excel this time but I can highly recommend "R" (even better with the RStudio environment) for statistical things. It's actually not complicated but probably even more unclear than Excel for a beginner.

[alphatier]

How much time does it take to get some decent results with "R" or Matlab?

[armory]

Is this all part of the AGW/CC agenda?

[dunno what to put here]

Originally Posted by armory

Is this all part of the AGW/CC agenda?

Haha.

[ncole1]

Originally Posted by nei

MATLAB for those charts, I switch back and forth between the two. Computer code is more efficient for large datasets, especially if doing similar graphs and processing. R is a good free alternative.

For those who like MATLAB, isn't OCTAVE a better free alternative?