Probability of precipitation
Improving the weather forecast with the aid of numbers
Weather is an incredibly complex phenomenon. Despite the use of computers,
satellites and skilled forecasters, it is still difficult to forecast future
precipitation. Rain or snow cannot always be predicted with a simple yes or
no. However, the probability of precipitation can help to gain a better understanding
of what the future weather holds. It offers us another way to view the precipitation
forecast.
To a large extent Environment Canada's weather service has always been in the
probability business. Forecasters have, at times, qualified their predictions
with subjective phrases like "scattered showers" or "possibility
of snow tomorrow". Forecasters can also express this type of probability
by numbers, adding a probability of precipitation statement to regional forecasts.
What is probability of precipitation?
Probability forecasts are a subjective numerical estimate that any point in
your forecast area will get measurable precipitation during the forecast period.
For example, a 40% probability of rain today means there are 4 chances in
10 of you getting wet today.
Basically, probability of precipitation forecasts allow the forecaster to express
his/her degree of belief in the likelihood of precipitation occurring. In
the regular forecast the forecaster might say "isolated showers",
but in the probability forecast he/she might predict a 20 percent probability
of precipitation. Again, if the public forecast says "there is a good
chance of precipitation tomorrow", the forecaster might follow this by
adding: 80 percent probability of precipitation.
Probability of precipitation and the weather forecast
ISOLATED SHOWERS THIS MORNING. CLOUDY THIS AFTERNOON. A FEW SHOWERS
TONIGHT. RAIN TOMORROW.
PROBABILITY OF PRECIPITATION 30 PERCENT TODAY, 60 PERCENT TONIGHT,
90 PERCENT TOMORROW.
As the above example shows, probability forecasts are part of the regional
forecasts for up to, three specific periods: today, tonight and tomorrow.
"Today" refers to the time period 6 am to p.m., "tonight"
refers to the time period 6 p.m. to 6 am, and tomorrow refers to the time
period 6 a.m. to midnight the following day. Probabilities are given in 10%
increments ranging from 0 to 100%.
How is the probability forecast determined for a particular region? The forecaster
studies the current weather situation, including wind and moisture patterns
and determines how these patterns will change with time. Factored into this
is the effect of terrain, long-term weather statistics and the character of
the precipitation (showery or continuous). In the case of showers, it is unlikely
that precipitation will occur at all points in the region, whereas continuous
rain is likely to be more widespread.
The procedures for producing a probability forecast do not differ radically
from those used in producing a regular forecast, but the forecaster does spend
more time checking all points in a region before determining the possibility
of precipitation.
The limitations of probability forecasts
Probability forecasts cannot be used to predict when, where or how much precipitation
will occur. For example, a 60% probability of snow today does not mean that
it will snow during 60% of the day. However, the probability figure does mean
that there is a 60% chance of a measurable amount of snow falling at that
location.
A person who did not encounter precipitation during the period would be tempted
to say the forecast probability should have been zero, while the individual
who did see precipitation during the same period would say the probability
should have been 100%. Statistically one cannot determine the reliability
of a single probability forecast. The reliability can only be verified after
a number of forecasts. A 30% reliability of precipitation forecast is reliable
if the same forecast was made on one hundred occasions and if it rained on
30 of those occasions.
A USER'S GUIDE TO PROBABILITY OF PRECIPITATION
| 0% |
No precipitation even though it may be cloudy. |
| 10 % |
Little likelihood of rain or snow: only 1 chance in 10. |
| 20% |
No precipitation is expected. |
| 30% |
If you go ahead with your outdoor plans, keep an eye on the weather. |
| 40% |
An umbrella is recommended. Make alternate plans for outdoor activities
that are susceptible to rain. Not a good day to pave the driveway.
Keep your fingers crossed! |
| 50% |
It's 50-50 on whether you get precipitation or not |
| 60% |
Want to water your lawn? The odds are favourable that Mother Nature
might give you some help. |
| 70% |
Consider the effect of precipitation on your plans for outdoor
activities. The chance for no precipitation is only 3 in 10! |
| 80% |
Rain or snow likely |
| 90% |
The occurrence of precipitation is a near certainty. |
| 100% |
Precipitation is a certainty. |
The weather office forecasts a probability of precipitation for your region
based on the information available at the time the forecast is issued. As
the time of the event nears and more information becomes available, the prediction
becomes more accurate.
It should be emphasised that probability forecasts are not meant to replace
regular weather forecasts. They simply supply a useful numerical addition
to the existing forecasts.
Using probability forecasts
Probability forecasts are more than a novelty or an interesting variation on
existing forecasts. They have considerable practical value and should allow
many people to make better decisions about weather sensitive activities.
To see how probability of precipitation may be used in making decisions, consider
a contractor who has to decide whether or not to pour some concrete for a
job. Suppose the probability of precipitation is 40%. The contractor has to
calculate the costs of re-doing the job if it rains and, if he/she goes ahead,
the expectation is that the risk is 40% of that cost. Compare this to the
costs of delaying the job until the next day. If the costs of delay is less
than the risk of going ahead, the best decision would be to delay. If, however,
the risk is less than the cost of delaying, then it is worthwhile to go ahead.
A good estimate of the probability of precipitation can help reduce costs
of weather-sensitive businesses.
Probability forecasts are also useful to farmers making crop decisions, to
promoters organising outdoor sporting events and to your own personal picnics
or skiing trips.
Generally, probability of precipitation forecasts aim to establish better communications
between the weather service and the public. Adding numbers to supplement the
existing forecasts provides Canadians with more useful weather forecasts.
For further information contact:
Environment Canada
Atmospheric Environment Service
Weather Services Directorate
4905 Dufferin Street
Downsview, Ontario M3H 5T4
Created :
2002-08-23
Modified :
2002-12-19
Reviewed :
2002-12-19
Url of this page : http://www.msc.ec.gc.ca
/cd/brochures/probability_e.cfm
The Green LaneTM,
Environment Canada's World Wide Web Site.
| 
