History
Surface
wetness or leaf wetness
Defintions
of surface wetness
Use
of multiple sensors
Frequency
distribution of surface wetness
Next
section measurement of surface wetness
Return
to table of contents
Surface wetness is an important variable for the prediction of many
plant diseases. Click
here for a brief history of SWD research.
1.1 Surface wetness or leaf wetness?
In general plant pathologists use the term "leaf wetness" to describe moisture in a canopy because the predominant plant organ is a leaf and because most pathogens infect the leaves. But in this context the term "leaf wetness" is not general enough. Pathogens such as Erwina amylovora, the causal agent of fireblight, infect the blossoms of plants, and moisture may remain in the calyx of a flower long after water has evaporated from leaves. In this publication it is proposed that "surface wetness" is a better term for defining the general wetness of a plant canopy. Surface wetness has been used by several authors in the past when discussing leaf wetness (Deshpande et al., 1995, Huber and Gillespie 1992). In this paper, moisture on any part of a plant will be defined as "surface wetness," which, of course, encompasses "leaf wetness."
1.2 Definitions of surface wetness duration
Surface or leaf wetness duration has been a poorly defined term in the plant pathology literature. Most commonly the term refers to a no/yes response of an electronic or mechanical sensor (Huber and Gillespie 1992). Rather than base the definition of surface wetness on an instrument, it is more appropriate to define the variable in physical terms. The following definitions were proposed by Magarey et al. unpublished # 1 & 2).
Surface
wetness
Canopy
surface wetness
Surface
wetness hour
Surface
wetting period
Surface
wetness duration
Return
to table of contents
Surface Wetness - The presence of water that is visible to the human eye on a material surface. Different surfaces may be defined, for example leaf, fruit or canopy.
Unit: Yes (1) or No (0) at different sampling frequencies.
At first it appears counterintuitive to classify SWD as a binary variable.
An alternative would be to express the proportion of the hour wet. However,
if the interval is partially wet it is not clear what part of the interval
was wet or dry.
Canopy Surface Wetness - The presence of water that is visible to the human eye on 10% or more leaf surface area in a canopy.
Unit: Yes (1) or No (0) at different sampling frequencies.
Since small traces of water may persist in the canopy for extended periods
it is useful to include a wet surface area threshold, Wt, below
which the canopy is considered to be dry. In this volume a value of 10%
was chosen to represent this threshold.
Surface Wetness Hour - An hour is considered wet if water is present for 12 minutes or more on a given material surface.
Unit: Yes (1) or No (0) at an hourly sampling frequency.
Hours are a suitable unit because weather data is commonly collected
on an hourly basis. The use of the 0.2 hour (12 minute) threshold assigns
hours that were partially wet to be considered wet, thus minimizing underestimation
of the variable. The disadvantage of a binary value is that it creates
errors that are on the average 0.5 hours, which is not a negligible error.
For example, infection may occur in less than 3 hours for Plasmopara
viticola, the causal agent of grape downy mildew (Orlandini et al.
1991). In instances where this is important a sub-hourly binary variable
should be considered.
Surface Wetting Period - The length of time during which a surface remains wet.
Unit: time (minutes, hours, days, etc.)
In many instances SWD data has been analyzed on the basis of daily totals
of wet hours (Gleason et al. 1978, Weiss et al. 1989). This
can be a useful means of comparing different sensors or summarizing SWD
data. However in plant pathology, the length of individual wet and dry
periods may be of importance (Huber and Gillespie 1992). In this volume,
SWD is analyzed by individual wetting periods and not daily summaries.
An example
of how wetting periods are calculated and a fortran
program for this estimation are available.
Surface Wetness Duration - The sum of individual surface wetting periods during a defined time interval.
Unit: time (minutes, hours, days, etc.)
Use of multiple sensors
When multiple sensors are deployed in the canopy it can become difficult
to interpret the output. Magarey et al. (unpublished #2) recommends
the following procedure.
When multiple sensors are positioned in the canopy, the relative wet area can be assumed to be equal to the proportion of the sensors that are wet. This assumption has greater validity with increasing sensor number. In this study, each SWD sensor recorded the proportion of the hour that was wet. If the proportion of the hour wet was greater than 0.2 then the hour was given a value of one, otherwise it was assigned the value of zero. For each hour these binary values from all sensors in the canopy were averaged. If the average was greater than the canopy surface wetness threshold (0.1) then the canopy surface wetness hour was equal to one, otherwise it was zero. An example of the calculation of wetting periods from raw sensor data is provided below. Wetting periods can be calculated with the aid of a Fortran program (Microsoft, Redmond, WA, 98052-6399) which can be dowloaded. In this example program inputs are temperature, humidity, rain and three leaf wretness sensors but obviously the program can be easily adapted .
Return
to Protocol discussion
Return
to table of contents
1.3 Frequency distribution of surface wetness duration
No information is available here at present.
Next
section measurement of surface wetness
Return
to table of contents