The use of the catch per unit effort (CPUE) as an index of abundance usually requires a standardization process consisting of isolating all those exogenous factors from temporal variations in abundance from the CPUE time-series. These exogenous factors include those generated by modifications in fishery vessel efficiency, variations in fishing strategies, and environmental fluctuations. The selection of the latter has been considered to be one of the most difficult, arbitrary, and poorly documented stages since the environmental effects vary on different temporal scales in autocorrelated and non-random manners, influencing the CPUE through a cause-effect process. Transfer function models (TFM) were constructed to describe statistically the cause-effect relationship between two time-series and herein we propose that TFM are a valid tool for: i) discriminating environmental effects that influence the CPUE and ii) describing how these effects should be included in a generalized lineal model (GLM). We analyzed the Antarctic krill CPUE from August 1989 to July 1999, and as possible causal effects, the Antarctic Oscillation Index (AOI) and atmospheric pressure at sea level (APSL). TFM shows that the APSL, with an annual lag (APSL12), influences the CPUE of Antarctic krill, whereas the AOI did not have a significant effect. The use of APSL12 in the GLM increased the explanation of the deviance by 31% as compared with the APSL with no lag. We concluded that TFM constitute a promising tool for including environmental effects in the standardization of the CPUE that would result in less biased and more accurate indexes of abundance.