The data sets we used have limitations. For example, most rain-gauge accumulations are derived from 24-hour reports manually taken from scattered locations along the coastal states. This does not allow to know rates for shorter time periods such as hours or minutes. In general, during the study period, there are few weather stations that included wind sensors to have more accurate measurements of the storm intensity and structure. The population counts, from the Mexican government, are performed every 5-10 years but we were not able to determine the population distribution at the time of every landfall event or how this distribution changed just after the event. Another challenge was to determine the number of affected people and deaths directly related to each tropical cyclone. Here, we decided to apply a reliable data base (EM-DAT) that compiles information from various sources including government agencies, insurance companies, research institutes and press agencies.

Thanks for your comment and interest in our poster. With respect to tropical cyclones affecting the west coast of Mexico, the study is based on a particular set of 10 events that caused the largest impact on the population from 1970 through 2014. We used the Oceanic Niño Index (ONI) to identify El Niño, La Niña and neutral phases. Most of these tropical cyclones developed during El Niño (6 out of 10 times) or neutral (3 times) conditions. I believe that the poster by Dominguez and Magaña (the composites in their Figure 3) states that the tropical cyclone formation in the Eastern North Pacific (ENP) is influenced by warm ENSO phases during the period 1979-2009. The corresponding cluster has tracks moving westward while some of them made landfall in Mexico, as seen in their Figure 1B. The ENP clusters (their Figure 1A-D) seem to have more members (with both landfalling and non-landfalling tracks) which leads to results that may not be consistent with our conclusion. Also, note that the SST anomalies are from a daily basis computation (the cyclogenesis day) while the ONI is defined for a three-month running mean within a region bounded by 5⁰N-5⁰S and 120-170⁰W that is partially seen in their Figure 3.