Comment on "Recent changes in climate extremes in the Caribbean region by Peterson et al. (2002)
Stone, Reynold J.
In a recent study, Peterson et al. (2002) analyzed daily temperature and rainfall data and concluded that “the climate of the Caribbean is changing” and that “these changes generally agreed with what is observed in many parts of the world”. We critically examined the study due to suspected methodological flaws by focusing on the most widely publicized conclusion of the study - the number or percent of warm days exhibited an increasing linear trend during the period 1958-1999. We discovered two major flaws, namely, an inadequate sampling procedure and the inappropriate use of ordinary least squares regression analysis for detecting trend. We therefore concluded that these two major flaws invalidate the claim of an increasing linear trend in the number or percent warm days in the Caribbean from 1958 to 1999.