Loading...

Development of a Model for Monitoring and Prediction of Drought under Climate Change at Watershed Scale

Abbasian, Mohammad Sadegh | 2020

457 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 52638 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Abrishamchi, Ahmad; Moghim, Sanaz
  7. Abstract:
  8. The objective of this dissertation is to develop a model to monitor and project the long-term changes in meteorological drought under climate change at watershed scale. In this model, drought is defined based on joint precipitation-temperature values since negative correlation between precipitation and temperature implies that drier periods are often warmer, and therefore, the consequences of drought are more severe compared to low-precipitation periods with mild temperature. This is of particular importance considering global warming. To quantify drought, an indicator called precipitation-temperature deciles index, which is an extension of precipitation deciles index, is introduced. Copula is proposed to model precipitation-temperature dependence structure in the index. Principal component analysis is proposed to cluster stations within a basin such that homogenous stations in terms of temporal variability are identified. To project precipitation and temperature at homogeneous stations from the outputs of climate models under climate scenarios, a class of regression models called multilevel (or hierarchical) model is proposed as a multi-site downscaling model. A multilevel model decomposes the variability of precipitation/temperature into within-station and between-station variability, and hence, it can be used as a regional downscaling model. In addition, Bayesian model averaging is used to calculate the mean of models’ projections. The application of the methodology to Lake Urmia basin shows that all nine climate models project that temperature will increase constantly in the future (2020-2040, 2040-2060, and 2060-2080) relative to the base period (1985-2005) under both RCP4.5 and RCP8.5 scenarios. However, the sign of precipitation change differs among models and scenarios. Based on the mean of models’ projections, temperature will increase by 1.2 C on average during the two following decades while precipitation will not change significantly. As a result, the frequency of “extremely dry/hot” and “dry/hot” months will increase by 3.5 percent relative to the base period. Instead, the frequency of “normal” and “wet/cold” months will decrease by 3.5 percent
  9. Keywords:
  10. Copulas ; Downscaling ; Clustering ; Drought ; Climate Change ; Precipitation ; Urumieh Lake

 Digital Object List

 Bookmark

No TOC