A friend of mine recently commented on how two years of data shows a decided cooling trend. We must be careful to remember the difference between weather, which is what occurs on any given day, week, month, year, and even 11-year sunspot cycle, and climate, which is what occurs over the long haul.
Furthermore, statistics being what it is, one or two data points mean nothing. Furthermore, the answer to the question, “How many data points are enough?” depends both on what you’re trying to measure and the nature of the data itself.
If you know you’re measuring a straight line, two data points are sufficient to describe the entire line.
If you know you’re measuring a parabola, and you know the parabola’s orientation (axis), two points are again sufficient. If you don’t know its orientation, you’ll need three points.
If you’re conducting an exit poll at a precinct, measuring whether people are voted for candidate A or Candidate B, and no write-ins were allowed, you need to pick a Confidence Level, say, 99%, a Confidence Interval, say, +/- 3 points, and the population size, say, 35,000 people in the precinct. The answer is a sample size of 653. However, that’s not all, as you need to ensure the respondents are randomly selected throughout the voting period. The largely liberal news organizations failed to take this into account when they launched their glowing pro-Hillary polls in the 2016 election.
When you’re talking about climate, however, the samples for each location need to include temperature, humidity, pressure, precipitation types and amounts, cloud types and cloud cover, and solar irradiance on the ground for at least 24 times each day, multiplied by every day for decades — at least thirty years worth, but preferably about 300+, then, multiply times thousands of locations around the world. You also need to measure solar irradiance in space i.e. the Sun’s output, and we’ve had access to that information only over the last 40 years. Finally, we need to correlate the irradiance with sunspot activity and discount the effect of sunspot variability, which can last as much as a century.
In all, there’s at least 16 pieces of variable information to be recorded at least hourly at each location, along with at least 12 pieces of constant information for each location.
For each location, that comes to 140,160 pieces of variable information each year, times tens of thousands of locations.
The best locations for this information are airports. According to the Airports Council International (ACI) World Airport Traffic Report, there are currently 17,678 commercial airports in the world. Most of these report their current conditions to one of several database repositories.
The major problem with the IPCC reports, however, is that they’re approach is rather simplistic. They often don’t even know what information to ask because they’re largely tied to the weather model, rather than a physics model. There are a number of relevant variables of which they either completely discount or have never even heard.
Local and surrounding terrain features, for example, significantly impact the readings. These “anomalous terrain features” can be mathematically described with via a centroid location, elongation factor, distance, and direction. Winds blowing over a mountain range 200 miles upwind during humid weather are likely to experience more cooling due to cloud formation than they are during dry weather. Similarly, weather stations located near a body of water are affected quite differently when the winds are onshore vs offshore. Even absolutely identical air masses located 500 miles distant will arrive in Kansas bearing quite different properties on a perfectly clear day throughout the entire U.S. depending on whether the air mass traveled up from low-lying Texas, down from the northern latitude Dakotas, or west over mountainous Colorado.
The same is true for ocean data. “Mean oceanic surface temperature,” while a good metric, is woefully void of the entire story, as oceans have basins and mountain ranges, too, and even slight shifts in currents can vary “ocean weather” significantly.
Then there’s the mudstream media’s “97% of climatologists agree” meme. It’s more than a meme, however, as pro-AGP (anthropogenic climate change) forces are now creating videos demonstrating how 97% of climatologists agree…
…while ignoring the reality that their agreement originates from a single errant paper that was picked up by mudstream media itself and spread like wildfire.
New York Times bestselling author Alex Epstein, founder of the Center for Industrial Progress, reveals the origins of the “97%” figure and explains how to think more clearly about climate change in this YouTube video, below:
FYI, here’s the ear-tickeling but blitheringly idiot piece of PBS crap that started this conversation: