How best to interpret those precipitation forecasts
Friday, April 4, 2014
Understanding more about the probabilities and types of rainfall your area may receive can play a big part in your farming decisions. Precipitation is always a gamble and it is vital to know the odds
by PHIL CHADWICK
Precipitation is typically the most important weather element for farmers. It is important that they receive and understand the best precipitation forecast.
Typical public forecasts contain spatial and temporal descriptors that are meant to communicate the character of the precipitation. Chance, isolated and frequent are just a few of the "weasel" words that have been used over the years to convey the character and likelihood of precipitation. Saying there is "a chance of a passing shower" generally confuses clients as they try to relate the words to their own understanding and experiences. There are similar disparities within the meteorological community. We can do better than this. What is the solution?
The probability of precipitation (POP) is based on the laws of chance. First introduced in Canada in the 1980s, POP was already well established in the United States in the 1960s. Simply put, the probability of precipitation in a weather forecast is an estimate of the chance of a measurable amount of rain or snow falling anywhere in a given forecast region over the forecast period. The probabilities are given in 10 per cent increments.
In Canada, we never use 50 per cent, which would be equivalent to tossing a "loonie." Any POP of 40 per cent or less means you can gamble on not getting wet. Any POP of 60 per cent or more and the odds are that you will get wet. POP has nothing to do with the precipitation type, rate or accumulation more than a trace amount. It is a probabilistic and not a deterministic forecast.
The forecaster typically uses POP to express a combination of degree of confidence that precipitation will occur and the areal coverage of that precipitation over the forecast region within the forecast period. The best way to interpret the POP forecast is that "chance of rain 40 per cent" means there is a 40 per cent probability that at least a trace amount of rain will occur at any given point in the area within the stated time period. If there were 10 identical weather patterns with the same perfect forecaster and the identical forecast, you would get wet four times out of the 10.
There are two main precipitation event types to consider – convective and stratiform.
Convective showers and thunderstorms are notoriously challenging to predict. Typically, the meteorologist can be fairly certain these small-scale events will form, but the timing and the areal extent is unclear. POP is typically used to convey the uncertainty with these warm season events.
Precipitation amounts with convection can vary wildly from nil on the edge of a shower to 50 millimetres or more under a strong thunderstorm. This is a huge concern for farmers.
Stratiform precipitation is associated with much larger, more slowly moving low-pressure areas. Stratiform events are characterized by extensive horizontal development, as opposed to the more vertical development characteristic of convection. Stratiform precipitation covers large areas but exhibits little vertical development. The associated precipitation is relatively continuous and uniform in intensity (for example, steady rain versus rain showers).
Uncertainty is only significant on the outer edges of the storm. POP is also used to convey any uncertainty on the development and track of the low-pressure area. Stratiform precipitation amounts are much more conservative in nature and show less spatial and temporal variance.
By itself, POP is useful, but most farmers want to know how much precipitation can be expected. They would really like to know this for the next two weeks, which is about the limit for any skilful weather predictions. The application of ensemble forecasting can provide such information.
The North American Ensemble Forecast System (NAEFS) is a joint project involving the Meteorological Service of Canada (MSC), the United States National Weather Service (NWS) and the National Meteorological Service of Mexico (NMSM). NAEFS was officially launched in November 2004 and combines state-of-the-art ensemble forecasts developed both at the MSC and the NWS. When combined, the grand ensemble can provide weather forecast guidance for a one-to-14-day period that is of higher quality than that based on either set of ensembles alone.
It allows the generation of a set of forecast products that are seamless across the national boundaries between Canada, the United States and Mexico. The list of products continues to grow to better meet the needs of specific clients. It is to be hoped that farmers are in this client list and using this information.
The North American Ensemble Forecast System (NAEFS) products can be found at http://weather.gc.ca/ensemble/naefs/produits_e.html.
The EPSgrams (Ensemble Prediction System meteorological-grams) found at the above link contain a precipitation graph in addition to graphs of temperature, wind and cloud cover for locations across North America. The results from all of the ensemble forecasts are plotted by time along with coloured lines for the deterministic control member from three different models of the atmosphere. If the spread in this "whisker" plot is small, you can have more confidence that the prediction has value. If the whiskers are really long, then the atmosphere has yet to decide which path it will follow as it evolves. See Figure 1.
Whisker plot explained
The Ensemble means and standard deviation charts (http://weather.gc.ca/ensemble/naefs/cartes_e.html) also include a precipitation map for North America. A small sample of such a forecast is below. See Figure 2.
The contours are the ensemble mean predicted precipitation contoured at 1, 5, 10, 20, 50, 75 and 100 millimetres per 24 hours. The 24-hour day for this ensemble mean prediction can be whichever day is selected out to 14 days into the future. The colours reveal the deviation from the contoured value with the brightest colours reserved for the largest deviations and uncertain predictions. The map gives the spatial and temporal extent of the precipitation and the uncertainty in the predicted amounts.
Maps of probabilities of occurrence of several weather events (http://weather.gc.ca/ensemble/naefs/produits_e.html) include a series of maps for precipitation. See Figure 3. Ensemble POP by accumulation threshold and total accumulation is the particular product that could be of use to farmers. It would be worth the effort to learn how to fully interpret and use the information. Plus, it is free.
The probability of the contoured area receiving more than the threshold amount of precipitation (see table on the right) during the selected 24-hour period is displayed in one series of charts. The colour within the contoured area corresponds to a probability that the area will receive more than the threshold amount of precipitation within the 24 hours selected. This colour and probability are displayed in the legend on the right side of the map.
Another series of maps displays the probability of precipitation accumulation for a series of thresholds (see map above) during any specific forecast period as selected from the start/end menu. Both tables are selected from the same drop-down menu. Do not be alarmed by the vast amount of information. With a bit of practice, the maps can be most helpful.
(Note: See Figure 4. I copied the above drop-down table from the website. These are the meteorological quantities that are included in the Probabilistic Products by period for NAEFS.)
Farmers make important business decisions every day. Weather information is vital in the business of modern farming. It is important to get the best information and use it. Precipitation is always a gamble and it is vital to know the odds. BF
Phil the forecaster Chadwick has been a professional meteorologist since 1977, specializing in training, severe weather and remote satellite and radar sensing.
