Forecast of the Global TEC by Nearest Neighbour Technique

We propose a method for Global Ionospheric Maps of Total Electron Content forecasting using the Nearest Neighbour method. The assumption is that in database global ionosphere maps spanning more than two solar cycles, one can select set past observations have similar geomagnetic conditions to those current map. ionospheric condition be expressed by linear combination seen past. average these leads common components being preserved and not shared several reduced. based on searching historical dates closest map as prediction correspond time shifts horizons. In contrast other methods machine learning, implementation only requires distance computation does need previous step model training adjustment each horizon. It also provides confidence intervals forecast. has been analyzed full years (2015 2018), selected days 2015 2018, i.e., storm non-storm performance system compared with CODE (24- 48-h forecast horizons).