Laserfiche WebLink
<br />D~,~ ~?9 <br /> <br />APPENDIX A <br />EVAPORATION PAN FILLING ERROR <br /> <br />Low evaporation values were found to cause <br />most of the noise in the suppression/change in <br />temperature relationship. This suggested that an <br />analysis was needed to determine the accuracy of <br />the evaporation pan measurement. Although the <br />evaporation pan measurement does not have high <br />precision or accuracy. no studies dealing with this <br />topic were found. Evaporation values are subject to <br />errors from many sources including those inherent <br />to the measurement technique. <br /> <br />Most of the errors will be site specific. The <br />following list gives an idea of the possible <br />difficulties which may exist at a station. <br /> <br />1. Location of the obstacle that may shade or <br />shelter the pan from the sun and wind. <br /> <br />2. Pan must be fenced to keep out animals, <br />bul birds will still cause errors. <br /> <br />3. Pan and water surface not level. <br /> <br />4. Refill water quality should be high, <br />otherwise salinity will increase in the pans and <br />influence evaporation. <br /> <br />S. Algae growth in the water will change the <br />pan albedo. <br /> <br />b. Operator error in measurement: from a <br />change in operator. operator's estimate in chang- <br />ing wind or light conditions, and operator's ability <br />to consistently match the water level to the point <br />elevation. <br /> <br />7. Surface tension may cause a difference in <br />the reading, depending on whether water is being <br />added (dry point) or subtracted (wet point) from <br />the pans. <br /> <br />Since the first five problems were taken into <br />consideration and avoided. deviation in the pan <br />reading should be attributed to the remaining two. <br />An experiment was run to arrive at a value for the <br />magnitude of this error. <br /> <br />Starting with a full graduated cyliner with an <br />accuracy of t ml and a deficit in the evaporation <br />pan. the pan was repeatedly filled. The water was <br />poured back inlo the graduated cylinder after each <br />cycle (without any loss of water from the system) <br />until it was filled. The range of measurements from <br />this process were used to calculate the standard <br />deviation. The experiment was run on a cloudy day <br />with occasional gusts of wind. Forty readings were <br />taken. The standard deviation was converted back <br />to readings corresponding to the cylinder used to <br />till the evaporation pan normally. From this data, a <br />range of .23 inch was observed with the standard <br />deviation of .0 I inch. Using this value for the <br />standard deviation in pan measurements, the <br />percent error in daily readings is shown in Figure <br />19. <br /> <br />Using this standard deviation in the suppres- <br />sion equation. a range of suppression values was <br />established. By using lhe regression equation for <br />suppression, a change in temperature of 4.20C (the <br />mean station value) was computed to have a <br />suppression value of .30. <br /> <br />This suppression estimate was used to <br />calculate evaporation from a cooled pan (~T <br />4.20C) corresponding to normal evaporation values <br />from 0.0 to 1.0 inches. The two values of <br />evaporation with adjustments due to the standard <br />deviation were then used to calculate suppression <br /> <br />53 <br />