A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Probabilistic Prediction of Drought in Iran Using Homogenous and Nonhomogenous Markov Chains
https://orcid.org/0000-0003-2138-0973
The main objective of this study is to predict the transition probability of different classes of droughts in Iran. The daily precipitation data of 40 synoptic stations in Iran, for a period of 35 years (1983–2018) were used to access the main objective of this study. The effective drought index (EDI) was used to recognize the abundance of various drought classes in Iran. Using cluster analysis on the monthly values of EDI, Iran was divided into five separate regions. Homogenous and nonhomogenous Markov chains were used to extract four features, including the probability of occurrence of each class of drought severity, the average expected residence time in each class of drought severity, the average time of the first expected passage of different categories from drought to a wet class, and short-term drought prediction for five regions of drought in Iran. Results suggested that the probability of occurrence of various drought classes decreased with the increase of drought severity; therefore, the highest probability of drought pertained to the drought class. In all five regions under consideration, the homogenous Markov chain has demonstrated the continuity of the severe drought class more than other classes; in the nonhomogenous Markov chain, however, when the beginning month is April, the highest continuity of drought is noted in the severe drought class. The average time of the first expected passage to achieve a no-drought class increases with the severity of the beginning class of drought. Consistent with the nonhomogenous Markov chain formulation, the average time of the first expected passage from each drought class to the wet class is more significant in the beginning month of April than in other months. In regions where the precipitation regime is seasonal and limited to one or two seasons, as in Southeastern Iran, few diverse states of droughts are noted; as a result, drought predictions will be more uniform, and the continuity of their various drought classes will be far greater. However, in areas where their precipitation regime is not limited to a specific season and precipitation occurs all year long, as in the Southern Caspian Sea, the predictions are more diverse, and continuity of the various drought classes is less; thus, the accuracy of the nonhomogenous Markov chain predictions can be more significant for arid and semiarid regions.
Probabilistic Prediction of Drought in Iran Using Homogenous and Nonhomogenous Markov Chains
https://orcid.org/0000-0003-2138-0973
The main objective of this study is to predict the transition probability of different classes of droughts in Iran. The daily precipitation data of 40 synoptic stations in Iran, for a period of 35 years (1983–2018) were used to access the main objective of this study. The effective drought index (EDI) was used to recognize the abundance of various drought classes in Iran. Using cluster analysis on the monthly values of EDI, Iran was divided into five separate regions. Homogenous and nonhomogenous Markov chains were used to extract four features, including the probability of occurrence of each class of drought severity, the average expected residence time in each class of drought severity, the average time of the first expected passage of different categories from drought to a wet class, and short-term drought prediction for five regions of drought in Iran. Results suggested that the probability of occurrence of various drought classes decreased with the increase of drought severity; therefore, the highest probability of drought pertained to the drought class. In all five regions under consideration, the homogenous Markov chain has demonstrated the continuity of the severe drought class more than other classes; in the nonhomogenous Markov chain, however, when the beginning month is April, the highest continuity of drought is noted in the severe drought class. The average time of the first expected passage to achieve a no-drought class increases with the severity of the beginning class of drought. Consistent with the nonhomogenous Markov chain formulation, the average time of the first expected passage from each drought class to the wet class is more significant in the beginning month of April than in other months. In regions where the precipitation regime is seasonal and limited to one or two seasons, as in Southeastern Iran, few diverse states of droughts are noted; as a result, drought predictions will be more uniform, and the continuity of their various drought classes will be far greater. However, in areas where their precipitation regime is not limited to a specific season and precipitation occurs all year long, as in the Southern Caspian Sea, the predictions are more diverse, and continuity of the various drought classes is less; thus, the accuracy of the nonhomogenous Markov chain predictions can be more significant for arid and semiarid regions.
Probabilistic Prediction of Drought in Iran Using Homogenous and Nonhomogenous Markov Chains
https://orcid.org/0000-0003-2138-0973
The main objective of this study is to predict the transition probability of different classes of droughts in Iran. The daily precipitation data of 40 synoptic stations in Iran, for a period of 35 years (1983–2018) were used to access the main objective of this study. The effective drought index (EDI) was used to recognize the abundance of various drought classes in Iran. Using cluster analysis on the monthly values of EDI, Iran was divided into five separate regions. Homogenous and nonhomogenous Markov chains were used to extract four features, including the probability of occurrence of each class of drought severity, the average expected residence time in each class of drought severity, the average time of the first expected passage of different categories from drought to a wet class, and short-term drought prediction for five regions of drought in Iran. Results suggested that the probability of occurrence of various drought classes decreased with the increase of drought severity; therefore, the highest probability of drought pertained to the drought class. In all five regions under consideration, the homogenous Markov chain has demonstrated the continuity of the severe drought class more than other classes; in the nonhomogenous Markov chain, however, when the beginning month is April, the highest continuity of drought is noted in the severe drought class. The average time of the first expected passage to achieve a no-drought class increases with the severity of the beginning class of drought. Consistent with the nonhomogenous Markov chain formulation, the average time of the first expected passage from each drought class to the wet class is more significant in the beginning month of April than in other months. In regions where the precipitation regime is seasonal and limited to one or two seasons, as in Southeastern Iran, few diverse states of droughts are noted; as a result, drought predictions will be more uniform, and the continuity of their various drought classes will be far greater. However, in areas where their precipitation regime is not limited to a specific season and precipitation occurs all year long, as in the Southern Caspian Sea, the predictions are more diverse, and continuity of the various drought classes is less; thus, the accuracy of the nonhomogenous Markov chain predictions can be more significant for arid and semiarid regions.
Probabilistic Prediction of Drought in Iran Using Homogenous and Nonhomogenous Markov Chains
J. Hydrol. Eng.
Mahmoudi, Peyman (author) / Rigi, Allahbakhsh (author)
2023-05-01
Article (Journal)
Electronic Resource
English
Simplified staged-homogenous Markov model for flexible pavement performance prediction
| Taylor & Francis Verlag | 2016
Probabilistic Assessment of Drought Characteristics Using Hidden Markov Model
| Online Contents | 2013
Probabilistic Assessment of Drought Characteristics Using Hidden Markov Model
| British Library Online Contents | 2013
| Taylor & Francis Verlag | 2017