As you move from left to right on this graph it is obivous that the data is normalized. There just happen to not be as many samples from below -50 degrees in the displayed data pull. This normalization implies that the closer you are to 0 degrees the higher Max Temps are.

One thing that stands out right away about this graph is the amount of clustering around -5 though 0 degrees and at 40 through 60 Latitude degrees. I have more high Humidity samples than lower. However, it does not appear that there is an obvious high humidity trend around the equator.

The distribution of points in this graph is fairly spread out. There are clear lines at Cloudiness (%)'s: 0, 20, 75 There is possible evidence that cities closer to the equator are less clowdy. There seems to be verry little clusters close to 0 degrees

Obviously we can't have negative windspead but I have more low windspeed samples than high windspeed samples. Moving from left to righ on the X axis does not indicate variation in latitude would cause higher windspead. This is another graph with a wide distribution of points.