Evaluation of the Wyoming Weather Modification Pilot Project (WWMPP)

An evaluation has been conducted of a snowfall enhancement project in the Medicine Bow and Sierra Madre mountains of southeast Wyoming. The Wyoming Weather Modification Pilot Project (WWMPP) project was undertaken to determine whether cloud seeding of Wyoming orographic clouds with silver iodide enhances the natural precipitation process to produce increased snowpack. The project design and background is described in detail in Breed et al. (2013) including the statistical design. The project evaluation includes a physical evaluation based on observations and modeling studies, as well as a statistical evaluation using a randomized cross over design in which two similar ranges are considered simultaneously. This approach follows National Research Council recommendations that any evaluation of cloud seeding include statistical, physical, and modeling components (NRC report, 2006). The randomized cross over statistical design specifies that one range is randomly seeded while the other one serves as a control. This an efficient evaluation design if the precipitation in the two ranges is correlated. Based on historical SNOTEL data, the daily snowfall between the Medicine Bow and Sierra Madre ranges is fairly well correlated (correlation coefficient of 0.5), leading to the adoption of this approach to statistically evaluate the project. The experimental design (Breed et al. 2008) specifies that seeding to take place when conditions in the two ranges meet certain criteria: 1) 700 mb temperature less than -8 °C (from sounding or model), 2) supercooled liquid water is present in both ranges (from radiometers or model), and 3) the 700 mb wind direction is from 315 and 260 degrees (from sounding or model). Snowgauge measurements using project gauges under these more restrictive conditions showed a correlation of 0.5 over the six years of the project, similar to that estimated from SNOTEL prior to the experiment, validating the cross over design approach. The statistic specified for the evaluation (Breed et al. 2013) is the Root Regression Ratio (RRR). This statistic takes into account controls located beyond the two ranges to account for natural variability. Breed et al. (2013) describes the design, instrumentation, approach to case calling, and proposed evaluation for the WWMPP. The WWMPP was conducted for six winters (2008 – 2013) and collected 154 Experimental Units (EUs) consisting of 4 hour of precipitation data at both target and control sites, which was reduced to 118 through quality control procedures. This study will presennt the primary and associated analysis of the WWMPP based on the 118 quality controlled four hour snowgauge precipitation data collected during the EUs.