Mon, 05 Feb 2018 15:00:46 +0000
UPND President Hichilema Hakainde yesterday attributed the current crop failure arising from rain deficit, to poor planning by the Patriotic Front, which he said had no leadership, because it failed to invest in sophisticated weather forecasting equipment that would have averted the situation, but instead in roads where money was stolen.
Once in power, he said, the UPND would invest in such sophisticated equipment that would forecast weather one year in advance accurately.
HH is lying
No such accurate forecast exists. Chuck Doswell a research meteorologist since the mid-1970s says: Weather forecasts are never perfect and it’s unrealistic to expect them ever to be. Furthermore, the weather can vary over relatively small areas, so a forecast that’s reasonably close in one part of a city may be much less accurate in another part. If we assume some sort of “tolerance” on the accuracy (say, having the forecast high temperature be within 2–3 degrees of the actual observed temperature), then a large majority of 24-hour forecasts are within that range of accuracy. As the “projection time” of the forecast increases – out to 48 hours, 72 hours, 96 hours, and so on – the accuracy decreases. After a period of 10–15 days, the accuracy of the forecast has fallen to the point
where a “climatology” forecast (i.e., the average weather for that location and date) is just as accurate. This 10–15 day weather predictability limit (predictability varies from day to day!) is not likely to change significantly as we learn more about the atmosphere – it’s inherent in the nature of the physics that governs the atmosphere. The combination of our inadequate data and our incomplete understanding, coupled with the nonlinear dynamics of the atmosphere, means that small errors in the forecast at the beginning can grow to the point where the forecasts essentially have no value over “climatology”. Not only is prediction of the atmosphere challenging, but it’s also challenging to predict the predictability!
The detailed concepts of “nonlinear dynamics” involve more than a layperson’s knowledge of mathematics, unfortunately. In linear systems, a small change produces a small effect, but in nonlinear systems, a small change can produce a large effect, under certain circumstances. This is inherently an unsolvable issue, barring some unforeseeable (at present) breakthrough.
The difference between a weather forecast and a climate forecast is quite significant. Climate is what we expect, and weather is what we get. When the weather forecast is at the predictability limit of 10–15 days, the appropriate weather forecast should be for the climatological average.
Weather is the condition of the atmosphere at a particular place over a short period of time, whereas climate refers to the weather pattern (statistics) of a place over a long period , long enough to yield meaningful averages.
