The social, economic, and environmental effects associated with extreme precipitation events are significant, driving the development of research on topics such as adaptation, mitigation, vulnerability, and climate change. The implications have been so substantial that the Intergovernmental Panel on Climate Change (IPCC) and the World Meteorological Organization (WMO) have issued reports that compile both scientific advancements and the losses related to hydrometeorological phenomena. Consequently, there are numerous studies that not only address the mechanisms associated with extremes but also those related to changes in global hydroclimatic patterns. These studies employ numerical tools such as Global and Regional Climate Models (GCM and RCM respectively), which enable the analysis of historical simulations to make projections of atmospheric behavior and inform decisionmaking. One of the mechanisms associated with changes in global precipitation patterns is the oscillation of Sea Surface Temperature (SST) anomalies. An anomaly refers to the difference between the climatological average of Sea Surface Temperature and the observed value at a specific point in time. Numerous studies have shown that this anomaly oscillates over different temporal scales and exerts various influences on global precipitation patterns, including droughts, hurricane cyclogenesis, changes in the width of the tropical belt, and extreme precipitation events, among others. For this reason, these anomalies are classified as Teleconnection Indices (TIs). Understanding precipitation patterns through Regional Climate Models (RCMs) and their link to TIs is essential for improving predictions of such phenomena. The Expert Team on Climate Change Detection and Indices (ETCCDI) has developed a set of Climate Extreme Indices (CEI) to characterize extremes in both precipitation intensity and the duration of wet and dry conditions. In addition, the Standardized Precipitation Index (SPI), endorsed by the WMO and adopted by several countries, has been widely used to quantify moisture conditions, allowing for the assessment of both extreme wetness and drought based on standardized numerical thresholds. In Mexico, research has demonstrated that Teleconnection Indices (TIs), such as the El Niño-Southern Oscillation (ENSO) and the Pacific Decadal Oscillation (PDO), are associated with droughts, extreme precipitation, and increased moisture contrasts between the northern and southern regions. These relationships are captured by various precipitation indices, including the Climate Extreme Indices (CEI) and the Standardized Precipitation Index (SPI).
This study aims to evaluate the ability of RCM simulations, forced with ERA-Interim data, to reproduce the spatial patterns of the temporal correlation between TI (ENSO and PDO) and SPI and CEIs over Mexico, using four RCM simulations for the period 1980 to 2012. To achieve this objective, (1) a climatological evaluation was conducted for both the precipitation from the simulations and the forcing data, as well as the SST from ERA-Interim, during the aforementioned period, (2) the spatial pattern of the temporal correlation between the TI and the SPI and CEIs calculated with ERA-Interim and the simulations were compared with observations.
The first part of this thesis examines the ability of ERA-Interim and the RCM simulations to emulate the spatial pattern of the temporal correlation between the TI and SPI. This was carried out by (1) evaluating the spatial and temporal representation of the SST using ERA-Interim and the climatological annual cycle of monthly accumulated precipitation from the simulations; (2) assessing whether ERA-Interim accurately represents the TI, and whether both the RCM simulations and ERA-Interim successfully reproduce the SPI24 and SPI60 patterns; and (3) studying the representation, of the spatial pattern of the temporal correlation between the TI and SPI24 and SPI60. The results indicate that ERA-Interim efficiently reproduces the characteristics of the SST and TI. Despite some overestimations present in both the simulations and ERA-Interim, the characteristics of precipitation are successfully reproduced. However, the calculation of SPI24 and SPI60 is underestimated both temporally and spatially. Finally, the spatial pattern of the correlation between ENSO and PDO with SPI24 and SPI60 displays the contrast between northern and southern Mexico. However, this pattern shows an underestimation in spatial variability.
The second part of this thesis assessed the RCM ability to replicate the correlation pattern between ENSO and the CEIs. The CEIs used in this study are Counting of Dry Days (CDD), Counting of Wet Days (CWD), maximum precipitation in one day (Rx1) and maximum precipitation in five consecutive days (Rx5). To achieve this objective, (1) an evaluation of the temporal characteristics of the monthly accumulated precipitation was conducted throughout the entire study period, as well as the temporal representation and phases of ENSO with SST from ERA-Interim; (2) the study evaluated the RCM simulations capacity to represent the spatial features of the CEI used; and (3) it analyzed whether the RCM simulations and ERAInterim reproduce the spatial pattern of the temporal correlation between ENSO and the four CEIs. The results of this section show that ENSO and its phases are reproduced by ERAInterim. The spatial representation of the CEIs, by both ERA-Interim and the simulations, shows overestimations in spatial variability and magnitude. However, it successfully represents the main spatial features observed. The simulations from Canadian Regional Climate Model version 5 (CRCM5) better represent the spatial pattern of the temporal correlation for certain indices (CWD and Rx5). However, the RCA4 simulation shows better representation for the correlation with the CDD index. Finally, in the seasonal representation of the correlation, the simulations from CRCM5 perform better in summer for the CDD and CWD indices.
In the final part of this thesis, the ability to represent the characteristics of the spatial correlation pattern between the PDO and the aforementioned CEIs was evaluated. Therefore, (1) ERAInterim dataset was evaluated to represent the temporal pattern of the PDO during the period considered in this thesis; (2) the representation of the climatological characteristics of the monthly precipitation and the seasonal representation of the CEI with ERA-Interim data and the simulations were studied; and (3) the spatial correlation pattern between the PDO and the CEIs was analyzed. The results for this part of the thesis show that ERA-Interim is capable of reproducing the main characteristics and phases in the temporal pattern of the PDO. For all datasets an overestimation was observed in the representation of the monthly precipitation, which is mostly attributed to the spring and summer seasons. The simulations from CRCM5 show a better fit in representing the CEI seasonally in the CWD, Rx1, and Rx5 indices, while RCA4 performs better for CDD. However, each simulation provides important information for each season, each index, and each region within Mexico. For the entire period (1980 - 2012), the spatial pattern of the temporal correlation does not show a defined pattern. Nevertheless, in the seasonal analysis, the correlation pattern shows that in spring, ERA-Interim and the RCM simulations capture the main spatial gradient described by observations. In winter, ERAInterim presents an overestimation; however, both ERA-Interim and RCM simulations manage to identify the spatial features of the correlation.
Overall, it was observed that both the ERA-Interim data and RCM simulations capture the main climatologically pattern of the temporal correlation. However, it should be noted that these datasets could provide better information depending on the season and the region considered within Mexico and within the specific time frame and TI. Therefore, due to this uncertainties the valuation most continue for different RCM and different GCMs.
| Date | 19 Dec 2024 |
|---|
| Original language | American English |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Annie Poulin (Supervisor) & Rabindranarth Romero-López (Co-supervisor) |
|---|