The traditional catch-can technique for measuring water depth application under sprinkler irrigation systems has the limitation of being unable to monitor the continuous variation of wind drift and evaporation losses (WDEL) under changing weather conditions. Such information is essential to better manage the agricultural water by improving the global water application uniformity under moving irrigation machines. Three parallel, long, impermeable water collection strips were constructed underneath a stationary linear move irrigation system to address this issue. The sprinkler discharge efficiency (SDE≈1- WDEL) was monitored over 5-min intervals (SDE5min) during Apr-Aug of 2014-on a bare experimental plot. Experiments were conducted on a discrete basis for about 1040h total in order to collect more than 11,600 SDE5min data points. It was found that the SDE5min is very dynamic and can experience abrupt changes up to 16.5% as a result of sudden changes in wind speed and direction. The maximum and minimum observed SDE5min during the study period were 97.5 and 73.6%, respectively. The difference between maximum and minimum SDE5min during a day/night cycle was controlled by solar radiation and maximum wind speed. The average hourly SDE (SDEhr) for data collected at predawn, morning, afternoon, and evening were 89.5, 87.7, 86.9 and 88.8%, respectively. Overall, the daily SDE was mainly controlled by air temperature and varied between 81.8 and 91.8%, indicating that ∼8–18% of the daily applied water was lost on average. A multiple linear regression model was developed to help predict the SDEhr as a function of weather parameters. The model indicated wind speed, temperature and relative humidity as the best explanatory variables and predicted the SDEhr with an absolute error just over 4%. Because the difference between maximum and minimum SDEhr over a full experimental day could be as high as 22%, this magnitude of error was considered to be acceptable for wind drift and evaporation loss estimation.