Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Recreational benefit estimates can be substantially improved by incorporation of data on attributes that are averaged over shorter periods of time so as to better represent within-season variability. This approach may be advantageous across a wide range of non-market valuation settings. We use angling survey data that includes bimonthly data on angler demand, water quality and fish weight in the Rotorua Lakes and compare our results with those obtained using attribute data averaged over the year. Estimates of marginal willingness to pay for catching fish based on annual data are 5 times higher than those based on bimonthly data. Likewise, the estimated welfare gain from a 1m increase in water clarity is 1.5 times higher in models using annual data. We find that within-season variability in recreational site attributes can have a statistically significant effect on welfare estimates. The effect of seasonal variation in site attributes has rarely been accounted for in travel cost random utility models of recreation. We conclude that appropriate treatment of temporal variability can assist environmental managers to design more cost-effective policies based on improved estimates of welfare gains.