Tuesday, November 14, 2006

Adaptation of a Meso-scale Rainfall Simulator for Turfgrass Runoff Investigations.

Mark Carroll1, David Funk2, Matthew Katsuleres2, Gregory Bell3, and Michael A. Kizer3. (1) 1112 H.J. Patterson Hall,, University of Maryland, University of Maryland, Dept Plant Science & Landscape Architecture, College Park, MD 20742-4452, (2) 1114 H.J. Patterson Hall, College Park, MD 20742-4452, United States of America, (3) Oklahoma State University, Oklahoma State University, 360 Ag Hall, Stillwater, OK 74078-6027

The Coody and Lawrence (U.S. patent 5,279,151) meso-scale rainfall simulator is frequently used to evaluate chemical runoff from agricultural crops. This simulator employs 2.7 m high stationary spinner plate type sprinkler heads, placed along two lateral lines, to deliver a continuous supply of droplets having a size distribution and impact velocities similar to that of natural rainstorms. The simulator as originally designed is portable, but is cumbersome to move. The inability to rapidly and frequently move this simulator make it incompatible for use in turf areas that are regularly mowed. Two modified Coody/Lawrence simulators, one in Maryland and another in Oklahoma, were designed and constructed to overcome this limitation. The Maryland design resulted in a rainfall simulator that could be broken down, moved to an adjacent plot, and set back up in about an about an hour by 4 people. Thus allowing several meso-scale rainfall simulation events to be replicated in one day. The Oklahoma modified design involved the creation of a semi-permanent dual Coody/Lawrence rainfall simulator unit. The Oklahoma design eliminated the need to move the simulator to perform routine maintenance operations around the plots. Both modified Coody/Lawrence simulators were field tested to evaluate the ability of the basic simulator design to uniformly deliver simulation rainfall within meso- plots. Field audit coefficient of uniformities for a 13 m lateral spacing ranged from 77 to 82 % which was similar to that predicted by irrigation software. Rainfall measured at 0.6 m intervals revealed that individual transect deviations as great as +/- 56% of the average amount applied should be expected inside 12 m wide plots when using a Coody/Lawrence rainfall simulator.