Although there are some advocacy in promoting
the further applications of predictive model (e.g.,
Rodriguez et al., 2007), most species predictive
mappings stopped after the prediction models were
compared and the best one singled out (e.g., Maynard
and Quinn, 2007). To our knowledge, our efforts
represented a successful application to fully use the
predictive model to obtain ecological distribution data
on the Fairy Pitta in its breeding range. Without further
affirmation, both predictive model and mapping cast
doubts about their accuracy and usability (Loehle and
LeBlanc, 1996; Travis, 1996). In our study, we showed
the applicability of the model by extensive field work,
and the probability to detect the Fairy Pitta was higher
within predictive areas than non-predictive areas.