The physics of pan evaporation

Abstract

Evaporation pans are traditionally used as physical means to quantify the evaporative demand of the atmosphere to assist irrigation scheduling and water resources management. Long-term pan evaporation data available worldwide may be useful for climate science investigations. However, basic understanding of the physics of pan evaporation is limited, and this thesis attempted to fill this knowledge gap by experimental monitoring and theoretical formulations. An instrumented US Class A pan (with bird guard) that is a replica of that used by the Australian Bureau of Meteorology (BoM) was installed at Canberra Airport (Australia) and monitored for three years (2007-2010). The experimental facility measured the water level, temperature (water surface, air, bulk water, pan wall), (short- and long-wave) radiation, wind speed, air vapour pressure and atmospheric pressure at sub-daily intervals. These data formed the basis for investigating the aerodynamics and energy balance of the pan under non-steady state conditions. The key outcomes are the following: A framework for quantifying vapour transfer was developed by coupling Fick's First Law of Diffusion with boundary layer theory. It sufficiently represents the mass transfer of the experimental pan under non-steady state conditions when water surface temperature data is available. This approach provides quantitative interpretation for the boundary layer thickness and air properties for the pan. Generic equations to quantify heat transfer (radiative and sensible heat) of the pan were formulated by considering the system geometry (solar and pan), surface properties (albedo and emissivity) of the pan water surface and the pan wall. They enabled better understanding of the relation between pan evaporation and lake and reference crop evaporation. By combining the aerodynamic and radiative formulations developed from the experimental pan, two versions of a new steady state model (PenPan-V2C and PenPan-V2S) of pan evaporation were developed and evaluated using high quality pan evaporation data in Australian continent. The PenPan-V2S model was found to be as robust as the PenPan-V2C model and suited to general applications because of its simplicity. The results presented in this thesis should be useful for better interpretation of the long-term pan evaporation records. It is hoped that it would assist the evaluation of the outputs of General Circulation Models (GCMs) and the studies of water availability in a changing environment.