A markov chain-based bias correction method for simulating the temporal sequence of daily precipitation

Authors: LIU, HAN - Wuhan University; CHEN, JIE - Wuhan University; Zhang, Xunchang; XU, CHONG-YU - University Of Oslo; HUI, YU - Wuhan University

Submitted to: Atmosphere
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/11/2020
Publication Date: 1/16/2020
Citation: Liu, H., Chen, J., Zhang, X.J., Xu, C., Hui, Y. 2020. A markov chain-based bias correction method for simulating the temporal sequence of daily precipitation. Atmosphere. 11(1):109-126. https://doi.org/10.3390/atmos11010109.
DOI: https://doi.org/10.3390/atmos11010109

Interpretive Summary: Climate model outputs cannot be directly used to drive hydrological and crop models due to biases in projected rainfall amounts, frequency, and temporal sequence. Many bias-correction methods are routinely used to correct the distribution of daily rainfall amounts as well as the number of rain days in a year, but they usually fail to correct the temporal sequence of rainfall occurrence. The rainfall sequence is important to simulate crop growth and surface hydrology such as flooding. To address this problem, this study presents a new hybrid bias correction approach for correcting both rainfall amounts and temporal sequence. Compared with the traditional bias correction methods, the proposed hybrid bias correction considerably improves the generation of the temporal sequence of rainfall occurrence considerably for 7 out of 10 stations. This approach should be useful to climatologists and hydrologists for simulating climate change impact at field and small watershed scales for programs such as the Conservation Effect Assessment Project (CEAP) and Long-term Agro-ecosystem Research network (LTAR).

Technical Abstract: Bias correction methods are routinely used to correct climate model outputs for hydrological and agricultural impact studies. Even though superior bias correction methods can correct the distribution of daily precipitation amounts as well as the wet-day frequency, they usually fail to correct the temporal sequence or structure of precipitation occurrence. To solve this problem, this study presents a hybrid bias correction method for simulating the temporal sequence of daily precipitation occurrence by combining a first-order two-state Markov chain with a quantile-mapping (QM) based bias correction method. Specifically, a QM-based method was first used to correct the distributional attributes of daily precipitation amounts and the wet-day frequency simulated by climate models. The sequence of precipitation occurrence was then simulated using the first-order two-state Markov chain with its parameters adjusted based on linear relationships between QM-corrected mean monthly precipitation and transition probabilities of precipitation occurrence. The proposed Markov chain-based bias correction (MCBC) method is compared with the QM-based method with respect to reproducing the temporal structure of precipitation occurrence over 10 meteorological stations across China. The results show that the QM-based method is unable to correct the temporal sequence, with the cumulative frequency of wet- and dry-spell length being considerably underestimated for most stations. The MCBC method is capable of reproducing the temporal sequence of precipitation occurrence, with the generated cumulative frequency of wet- and dry-spell lengths fitting that of the observation well. The proposed method also performs reasonably well with respect to reproducing the mean, standard deviation and the longest length of observed wet- and dry-spells. Overall, the MCBC method is capable of simulating the temporal sequence of precipitation occurrence, along with correcting the distributional attributes of precipitation amounts, which can be used with crop and hydrological models for climate change impact studies at field and small watershed scales.