An international intercomparison of micrometerological techniques and dynamic flux chamber methods applied to measure mercury fluxes was conducted from September 1 to 4, 1997, during the Nevada Study and Tests of the Release of Mercury From Soils (STORMS) in Reno, Nevada. Nine research groups from four countries met in the Steamboat Springs, Nevada Geothermal Area, to participate in the first international flux intercomparison ever attempted for mercury. The highly heterogeneous soil Hg concentrations and complex landscape within the study area (4 ha) were unfavorable for spatial intercomparison of Hg fluxes between the research groups. However, reliable and correlated Hg fluxes were measured between our micrometerological technique and a dynamic flux chamber method (r2=0.29), run side by side (5 m). Hg fluxes and their relationships with environmental factors were complex. After ~90 days of dry condition, a series of storm events impacted the site and increased the soil moisture from <0.5 to 6.6%. This appeared to promote strong Hg evasion during the transition period from dry to wet soil conditions. However, the subsequent relationship between soil moisture and Hg flux was significantly negatively correlated. Multivariate analysis was applied to extract the principal components (principal component analysis). Three principal components were extracted (explained up to 79% of the total variance) and discussed with respect to their environmental signification. Environmental conditions under southern wind sectors were optimal to promote Hg fluxes. Turbulence rather than Hg air concentrations seemed to be the main factor promoting the determined Hg fluxes during this study. ¿ 1999 American Geophysical Union