In hydrology, as in a number of diverse fields, there has been an increasing use of Artificial Neural
 Networks (ANN) as black-box simplified models. This is mainly justified by their ability to model
 complex non-linear patterns; in addition they can self-adjust and produce a consistent response when
 ‘trained’ using observed outputs.
 This paper utilises various types of ANNs in an attempt to assess the relative performance of existing
 models. Ali Efenti, a subcatchment of the river Pinios (Greece), is examined and the results support
 the hypothesis that ANNs can produce qualitative forecasts. A 7-hour ahead forecast in particular
 proves to be of fairly high precision, especially when an error prediction technique is introduced to the
 ANN models.