Improving accuracy of rainwater tank hydrologic yield estimation across Queensland

Abstract

This dissertation aims to increase the accuracy of rainwater tank hydrologic yield and water saving efficiency estimation throughout Queensland. Traditional methods of determining average annual hydrologic yield from average annual precipitation and water saving efficiency from seasonality indexation are reproduced using current daily observations from the Bureau of Meteorology (BOM). Limits with these methods are identified as reduced regression confidence and increased dependence on site parameters. An alternative method of performance indexation is presented based on the failure principles of a rainwater tank mass balance simulation. The Taylor’s Hyetology Index (THI) allows hydrologic yield and reliability to be determined independent of location or rainwater tank volume. THI provides higher regression confidence and reduces the current number of charts needed to represent Queensland from 90 to 9. 2520 unique simulations were conducted with Aquacycle to provide the data for state-wide trend discovery. These simulations are the result of placing a model unit block at eight sites being Birdsville, Brisbane, Cairns, Caloundra, Charleville, Mount Isa, Rockhampton and Townsville. The model unit block takes many forms defined by five parameter dimensions. The parameter dimensions include effective roof areas (75 m2, 150 m2 and 225 m2), garden irrigation areas (Nil, 125 m2 and 250 m2), nominal rainwater tank volumes (3 kL, 5 kL, 7.2 kL, 10 kL and 14.5 kL) and occupancies of 1 to 7. Data drill BOM daily precipitation and FAO Penmam- Monteith potential evapotranspiration data for the period 01/01/1970 to 18/06/2009 was used for each site