Potential of the hazelnut tree search–ELM hybrid approach in estimating yield and water productivity

Abstract Wheat plays a vital role in the food security of society, and early estimation its yield will be great help to macro-decisions. For this purpose, wheat water productivity (WP) by considering soil data, irrigation, fertilizer, climate, crop characteristics using novel hybrid approach called hazelnut tree search algorithm (HTS) extreme machine learning method (ELM) was examined under drip (tape) irrigation. A dataset including 125 irrigation meteorological data Mahabad plain located southeast Lake Urmia, Iran, used as input parameters for year 2020–2021. Eighty percentage were training, remaining 20% model testing. Nine different scenarios presented estimate WP. The efficiency proposed calculated with statistical indices coefficient determination ( R 2 ), root-mean-square error (RMSE), normalized error, criterion. Sensitivity analysis result showed that rainfall, moisture, variety provide better results modeling. There good agreement between practical values (field management data) estimated HTS–ELM model. also is very efficient selecting best combination = 0.985 RMSE 0.005. In general, intelligent methods can enable optimal economical use resources.