The most serious limitation of this study was the daily resolution of the water quantity data. Other impediments were a lack of a robust and flexible hydrologic modelling system, although it is possible that Rhessys would have been effective if hourly stream flow data had been available. Furthermore, more numerous TM images, covering different times of the year and a greater number of years in recent decades, would have enabled a more thorough analysis of land cover/use change with urbanization.

Therefore, future study relating urbanization to storm water quality and quantity in the Humber basin would benefit from:

MORE STREAM GAUGES
To allow for more effective monitoring of storm water in specific areas that are rapidly urbanized, in order to determine the nature of the correlation between such urbanization and stream flow / water quality.

HIGH RESOLUTION WATER QUALITY DATA
Data which measured a number of parameters consistently on an hourly basis would allow more precise correlations between urbanization and water quality impairment. This would also allow specific contaminants of interest to be studied with regard to urbanization.

HIGH RESOLUTION WATER QUANTITY DATA (MIN. HOURLY)
In order to record storm flows in the Humber basin, hourly stream flow measurements would be required at a minimum. This is particularly true of the highly urbanized portions of the watershed, where precipitation events result in a stream flow response that is of greater magnitude and short-lived.

ROBUST HYDROLOGIC MODELLING SYSTEM
We presume that Rhessys would work with better quality data. However, a robust modelling system would facilitate the study of correlations between stream flow and urbanization even with relatively low-resolution data, i.e. daily to bi-hourly.

MORE HYDROLOGIC DATA (SOIL WATER, SNOW MELT, EVAPOTRANSPIRATION)
A more complete picture of the hydrologic system would require information pertaining to not just stream flow, but all of the hydrologic stores in the system, including ground water, snow, etc.

MORE THEMATIC MAPPER SCENES - VARIOUS SEASONS, YEARS
In this study, we compared Thematic Mapper images which were remotely sensed in 1985 and 1992 at two different times of the year. This made direct comparison problematic. Access to a number of scenes from different years and seasons would facilitate comparison.