CR prepress abstract  -  DOI: https://doi.org/10.3354/cr01545

Prediction of climate variables by comparing the k-nearest neighbor method and MIROC5 outputs in an arid environment

Hamid Reza Golkar Hamzee Yazd, Nasrin Salehnia*, Sohrab Kolsoumi, Gerrit Hoogenboom

*Email: Salehnia61@gmail.com

ABSTRACT: The goal of this study was to compare the ability of the k-Nearest Neighbors (k-NN) approach and the downscaled output from the MIROC5 model for generating daily precipitation (mm), and daily maximum and minimum temperature (Tmax, and Tmin) (C) for an arid environment. For this study data from the easternmost province of Iran, South Khorasan was used for a period from 1986 to 2015. We also used an ensemble method to decrease the uncertainty of the k-NN approach. Although based on the initial evaluation MIROC5 had better results, we also used the output result of kNN alongside the MIROC5's data to generate future weather data for the period 2018 to 2047. NSE (Nash-Sutcliffe efciency) between MIROC5 estimates and observed monthly Tmax ranged from 0.86 to 0.92, and 0.89-0.93 for Tmin over the evaluation period (2006-2015). K-NN performed less well, with NSE between k-NN estimates and observed Tmax ranging from 0.54 to 0.64, and from 0.75-0.78 for Tmin. The MIROC5 simulated precipitation was close to observed historical values (-0.06 < NSE < 0.07); the k-NN simulated precipitation was less accurate (-0.36 < NSE < -0.14). For these arid regions, the k-NN precipitation results compared poorly to the MIROC5 downscaling. MIROC5 predicts increases in monthly Tmin and Tmax in summer and autumn and decreases in winter and spring and decreases in winter monthly precipitation under RCP4.5 over the 2018-2047 period of this study. This study showed that the k-NN method cannot be used for generating future data in comparison to the outputs of the MIROC5 model for arid environments.