In recent decades,the impact of climate change on natural resources has in-creased.However,the main challenges associated with the collection of mete-orological data include the presence of missing,outlier,or erroneou...In recent decades,the impact of climate change on natural resources has in-creased.However,the main challenges associated with the collection of mete-orological data include the presence of missing,outlier,or erroneous data.This work focuses on outliers detection in long-term climate data by using machine learning models.The study uses meteorological data collected over 40 years(1981-2021)from ten synoptic stations operated by Burkina Faso’s National Meteorological Agency(ANAM).The methodology is based on the use of 18 machine learning algorithms from the PyOD library,including prob-abilistic,linear,proximity-based,and ensemble models.Univariate and mul-tivariate analyses are performed.For the multivariate analysis,this paper fo-cuses on two key variables,maximum temperature and minimum relative hu-midity which consistently exhibit strong correlations across all stations.A ro-bust approach is adopted to optimize the detection of outliers,using thresh-olds based on extreme percentiles.The results show that models such as KPCA,LSCP,LOF,and Feature Bagging are best suited to capturing anomalies in complex time series.These results will contribute to more reliable climate analyses and improved modeling of extreme climate events in data-scarce re-gions.展开更多
文摘In recent decades,the impact of climate change on natural resources has in-creased.However,the main challenges associated with the collection of mete-orological data include the presence of missing,outlier,or erroneous data.This work focuses on outliers detection in long-term climate data by using machine learning models.The study uses meteorological data collected over 40 years(1981-2021)from ten synoptic stations operated by Burkina Faso’s National Meteorological Agency(ANAM).The methodology is based on the use of 18 machine learning algorithms from the PyOD library,including prob-abilistic,linear,proximity-based,and ensemble models.Univariate and mul-tivariate analyses are performed.For the multivariate analysis,this paper fo-cuses on two key variables,maximum temperature and minimum relative hu-midity which consistently exhibit strong correlations across all stations.A ro-bust approach is adopted to optimize the detection of outliers,using thresh-olds based on extreme percentiles.The results show that models such as KPCA,LSCP,LOF,and Feature Bagging are best suited to capturing anomalies in complex time series.These results will contribute to more reliable climate analyses and improved modeling of extreme climate events in data-scarce re-gions.
