Observations of dew amount using in situ and satellite measurements in an agricultural landscape Michael H. Cosh a, *, Erik D. Kabela b , Brian Hornbuckle c , Mark L. Gleason c , Thomas J. Jackson a , John H. Prueger d a USDA-ARS Hydrology and Remote Sensing Laboratory, Beltsville, MD 20705, United States b Savannah River National Laboratory, Aiken, SC 29808, United States c Iowa State University, Ames, IA 50011, United States d USDA-ARS National Soil Tilth Lab, Ames, IA 50011, United States 1. Introduction The study of dew has focused primarily on its presence and duration as an indicator of plant pathogen activity (Monteith, 1957; Royle and Butler, 1986; Sutton et al., 1984) because many plant pathogens require moisture for germination (Wallin, 1963). As a result, most modeling activities and sensor technologies have focused on occurrence and duration of surface wetness (Gillespie and Kidd, 1987; Wilson et al., 1999). Wilson et al. (1999) developed a model that simulates dew formation and duration throughout the leaf canopy. Magarey et al. (2005) provide an overview of the development of dew monitoring that indicates a preponderance of research on dew duration rather than quantity. As remote sensing of soil moisture has evolved from concept to satellites, research has begun to consider the full range of factors that can affect the observation, which includes leaf wetness resulting from dew-fall, distillation, and guttation (Monteith, 1957). The effect of dew on microwave signals used for soil moisture remote sensing has been mixed. Initial studies (Jackson and Moy, 1999; Wigneron et al., 1996) reviewed previous research utilizing passive microwave remote sensing and concluded that there would be little to no effect at the L-band. Conversely, Pinter (1986) showed that dew had a significant effect on canopy reflectance across a range of frequencies by using radiometers during a diurnal cycle. More recent studies (Hornbuckle et al., 2007, 2006) determined that L-band radiometers (those used for soil moisture remote sensing) are reactive to the presence of leaf wetness. Wood et al. (2002) determined that leaf wetness affected the relationship between backscatter coefficient of active micro- wave remote sensing and crop characteristics such as yield and biomass. The magnitude of influence that dew may have has yet to be determined, but first a methodology for estimating large-scale dew magnitudes is necessary. Estimating the quantity of dew temporally on a daily scale and spatially on a regional scale will be valuable in further research and potentially in operational practices (such as irrigation scheduling and fertilizer applications), not only for remote sensing, but for modeling. The necessary input for any study of the effect of leaf wetness on remote sensing requires an estimation of the leaf wetness within the vegetation canopy on a scale at which satellite Agricultural and Forest Meteorology 149 (2009) 1082–1086 ARTICLE INFO Article history: Received 21 May 2008 Received in revised form 9 December 2008 Accepted 15 January 2009 Keywords: Leaf wetness Dew Electronic sensors Corn Soybeans ABSTRACT Estimating the amount of water on leaf surfaces is an increasing concern for remote sensing and hydrology. Measuring the magnitude and spatial extent of leaf wetness events will provide useful information for water and energy balance modeling and remote sensing. As part of the Soil Moisture Experiments 2005 (SMEX05), the temporal and spatial characterization of leaf wetness over a heterogeneous agricultural domain was investigated. Leaf wetness sensors and physical measurements were collected from 15 June to 3 July 2005 in and around the Walnut Creek Watershed near Ames, Iowa, USA. Comparison of the results of the in situ leaf wetness sensor measurements and the physical sampling revealed a moderate correlation for both corn (Zea mays L.) and soybeans (Glycine max Merr.). Regression equations were developed to estimate leaf wetness quantity from these leaf wetness sensors and combined with a vegetation leaf area index map to produce a spatial leaf wetness product hourly during the experiment with an error of approximately 0.05 kg/(m 2 LAI). Using this strategy, future efforts in spatial hydrologic modeling and remote sensing would be able to incorporate quantitative estimates of leaf wetness amount in watershed scale studies using only in situ measurements. Published by Elsevier B.V. * Corresponding author. Tel.: +1 301 504 6461; fax: +1 301 504 8931. E-mail address: Michael.Cosh@ars.usda.gov (M.H. Cosh). Contents lists available at ScienceDirect Agricultural and Forest Meteorology journal homepage: www.elsevier.com/locate/agrformet 0168-1923/$ – see front matter . Published by Elsevier B.V. doi:10.1016/j.agrformet.2009.01.004