You have a sphere. How do you distribute sensors over the surface of the sphere so you can accurate measure the true average temperature of the sphere? If you

You have a sphere. How do you distribute sensors over the surface of the sphere so you can accurate measure the true average temperature of the sphere? If you have one sensor, you can't measure an average. Add a sensor on the other size and you have a better chance of measuring an average but still a lot of room for error. Now add 100 more. If they are evenly distribute over the surface of the sphere you're likely to get a reasonable approximation of average. If, however, all those sensors are cluster around a single pole, then you effectively get accurate of measurement comparable to when you only had a single sensor. With this all in mind how are the world's sensors whose data is the basis of this article arranged on Planet Earth and do we know how much that deviates from actual average temperature? The reason I pose this because the majority of people live in places with temperatures agreeable to the human body and many of the humanities sensors will be measuring these places but won't include what's happening in all the other areas of the world that are generally outside that livable range. Just something to think about when we talk about global averages.