LONG RANGE FORECASTING SERVICE

this step show the verification of foreacst precipitation issue from previous forecast in precipitation map in percent of average, performance index map forecast verifcation

Using the forecast map issue 3 to 5 years ago, to identify the good performance of foreacast

Climate variability and trends for the target seasons is generate with monthly precipitation data downloaded using script. These are graphs processed with Excel or R package (see practical guide for seasonal climate forecast step1)

Analog years are the previous years where the behavior of SSTs in ocean basins is close to that of the current year. Comparing the temporal variability and the SST values of the historical years (from 1980 up to date) with current year and the forecast for next three months (see practical guide for seasonal climate forecast step2).

The wet and dry years are identified on the basis of percent of average precipitation. Years with percent of average precipitation above a threshold (120 or 125% to be chosen) are considered as wet years. Years are considered to be years in which the percent of average precipitation are less than a threshold (75% or 80% to be chosen) are taken as dry years. The choice of thresholds depend on the basis of the climatology of the location (see practical guide for seasonal climate forecast step3).

It is the process of averaging SST anomaly data for wet and dry years. Related maps are used to identify SST patterns associated with wet and dry years, respectively. These patterns are compared with the current and expected evolution of SST anomalies leading to determination of whether or not the coming season can be close to either a wet or a dry year (see practical guide for seasonal climate forecast step4).

Cumulative precipitation and percent of average time series graph are generated with the current year, cumulative average year 75% and 125% of the average year and analog year. The Graph facilitated drought analysis at grids point or at station (see practical guide for seasonal climate forecast step5).

Statistical seasonal precipitation forecast is generated using a statistical package: Climate predictability Tool (CPT). Statistical seasonal precipitation forecast is based on relationships between precipitation as predictants and predictors (example: SST). Output products are probabilistic precipitation with 3 categories (Above average, Near average and Below average) (see practical guide for seasonal climate forecast step6).

Analysis on SST, temperature and precipitation maps downloaded (see practical guide for seasonal climate forecast step7)

Analysis on SST, temperature and precipitation maps downloaded (see practical guide for seasonal climate forecast step8)

Probability of exceeding a given threshold during the season.

Map showing the onset status (Observe/Forecased).

Map to plot the rainfall profils for all synoptic stations.