Take a large, almost round, rotating sphere 8000 miles in diameter, surround it with a murky, viscous atmosphere of gases mixed with water vapor, tilt its axis so it wobbles back and forth with respect to a source of heat and light, freeze it at both ends and roast it in the middle, cover most of its surface with liquid that constantly feeds vapor into the atmosphere as the sphere tosses billions of gallons up and down to the rhythmic pulling of a captive satellite and the sun.
Now, try to predict the conditions of that atmosphere over a small area within a 5-mile radius for a period of one to five days in advance.
Such is the plight of modern meteorology. While we have come a long way indeed, a popular criticism contends that anything predicted beyond 36 hours quickly loses veracity. While in art school, I wrote a script to download the atmospheric forecasting products for a certain amount of time, and then contrasted what actually happened against what was forecast.