Effects of the prewhitening method, the time granularity, and the time segmentation on the Mann–Kendall trend detection and the associated Sen's slope

Collaud Coen, Martine; Andrews, Elisabeth; Bigi, Alessandro; Martucci, Giovanni; Romanens, Gonzague; Vogt, Frédéric P. A.; Vuilleumier, Laurent

doi:https://doi.org/10.5194/amt-13-6945-2020

Articles | Volume 13, issue 12

https://doi.org/10.5194/amt-13-6945-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/amt-13-6945-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 13, issue 12

Research article

|

21 Dec 2020

Research article |

| 21 Dec 2020

Effects of the prewhitening method, the time granularity, and the time segmentation on the Mann–Kendall trend detection and the associated Sen's slope

Martine Collaud Coen, Elisabeth Andrews, Alessandro Bigi, Giovanni Martucci, Gonzague Romanens, Frédéric P. A. Vogt, and Laurent Vuilleumier

Model code and software

mannkendall/ Matlab: First release M. Collaud Coen and F. Vogt https://doi.org/10.5281/zenodo.4134619

mannkendall/Python: First release F. P. A. Vogt https://doi.org/10.5281/zenodo.4134435

mannkendall/R: First release A. Bigi and F. P. A. Vogt, https://doi.org/10.5281/zenodo.4134633

Short summary

The Mann–Kendall trend test requires prewhitening in the presence of serially correlated data. The effects of five prewhitening methods and time granularity, autocorrelation, temporal segmentation and length of the time series on the statistical significance and the slope are studies for seven atmospheric datasets. Finally, a new algorithm using three prewhitening methods is proposed in order to optimize the power of the test, the amount of erroneous false positive trends and the slope estimate.