Early nineteenth-century United States hurricanes: A GIS tool and climate analysis.

摘要

Hurricane climate research is based on data spanning the last 100 years or so. To better understand rare but potentially catastrophic hurricane events it is helpful to have longer records. Records from historical archives are available, but they need to be collated and edited. Efforts to collate U.S. tropical cyclone information from the first half of the 19th Century using a Geographic Information System (GIS) have been conducted in this research. The Historical Hurricane Impact Tool (HHIT) is based on Environmental Systems Research Institute's (ESRI) ArcView GIS 3.1. Statements concerning coastal and near-coastal impacts are reproduced within map callout boxes. The callout boxes point to the geographic location of the documented information. Map layers are used for different archival sources. The HHIT, which is available in hardcopy format and will be online in the near future via an internet map server, can be used by scientists, emergency managers, and the general public to better estimate the risk of a hurricane catastrophe.; The combination of NOAA's “Best-Track” data back to 1851 and the HHIT collated hurricane list back to 1801 provide an unprecedented look at U.S. hurricane activity since the beginning of the industrial revolution. This research also examines U.S. (major) hurricanes over four 50-year epochs, and then further examines regional trends in U.S. hurricanes. I find that seasonal distributions are similar across epochs, and that the earliest epoch contains the greatest ratio of major hurricanes to all U.S. hurricanes. I further divide each epoch into three separate regions, and note that hurricane landfalls in Florida and the East Coast region have an inverse relationship. Furthermore, I note that the relationship between climate variables such as ENSO, the NAO, the PDO, and U.S. hurricanes appears to be different in the first epoch (1801–1850) than in the other three epochs (1851–2000). The relationships noted are robust to changes in sample size. A physical explanation for the noted trend is presented in a later chapter. Finally, additional climate influences on U.S. hurricanes, such as volcanic eruptions and sunspots, are explored for effects on landfall counts.