D. Hoppmann Deutscher Wetterdienst, GF Landwirtschaft, AST Geisenheim, Kreuzweg 25, D - 65366 Geisenheim

tel: +496722996111, fax: +496722996144, e-mail: dhoppmann@dwd.d400.de

K.-P. Wittich Deutscher Wetterdienst, GF Landwirtschaft, Agrarmet. Forschung, Bundesallee 50, D - 38116 Braunschweig, tel: +495312520526, fax: +495312520545, e-mail: wittich@zamf.fal.de

1. INTRODUCTION

It is commonly known that weather plays a key role in plant epidemiology. In particular, leaf-surface wetness produced by dew, fog or precipitation is one of the most significant meteorological pest-promoting factors that trigger fungal and bacterial plant diseases and activities of insects, and that influence the effectiveness of pesticides and the uptake mechanism for gases deposited onto vegetation. Many phytopathological models use the leaf-wetness parameter in combination with other factors in order to assess the infection risk and pest severity, and to manage disease control activities in an efficient way (Schrödter a. Ullrich 1966, Werres a. Hindorf 1991, ARAUZ et al 1990).

One of the objectives of the Deutscher Wetterdienst is to analyze the leaf-wetness risk potential of the weather and to forecast, besides others, agricultural pests and diseases dependent on leaf wetness.

2. LEAF-WETNESS CALCULATION SCHEMES

A great number of measuring principles and construction techniques are available for the monitoring of leaf-wetness duration.

Fig.1a: Simulation of a capacitive sensor response (continuous line) with a leaf-wetness model (circles and boxes) for a dew night (experimental site at Dossenheim, 1 October 1992).
 

Some techniques use artificial surfaces that are more or less representative of the shape or dimension of the leaves in question; other techniques use electronic grid elements which can be mounted directly on the leaf surface. Barthakur (1985) developed a radiometric beta-ray gauge system whose count rate depends on the water layer thickness on the leaf surface.

Leaf-wetness duration is also assessed by microcli-mate models which are calibrated by electronic leaf-wetness measurements in or above canopies. The models distinguish between low crops, for example potato or wheat crops, and canopies with foliage-free bottom such as apple orchards and vineyards (HOPPMANN a. WITTICH 1997). In low crops leaf-wetness duration is modelled by a multi-layer SVAT scheme taking the soil moisture, canopy interception and the crop-dependent radiative transfer into consideration (BRADEN 1995). In contrast to this, for orchards it is assumed that the soil has no effect on the leaf-wetness duration. In this case the calculation scheme is restricted on the top leaf of an orchard (Wittich 1993). The leaf can form and evaporate dew according to its energy balance and the flow state. Leaf-wetness duration caused by rain is given by the duration of the rain period itself and, during the following rainless hours, by the lifetime of a water drop settled on the top leaf.  

Fig.1b: Comparison of measured and modelled leaf-wetness durations on top of an apple orchard and of a vineyard (Dossenheim, April - October 1992; Geisenheim, September 1995).

Fig. 1a shows a dew simulation on top of an apple orchard which is compared with the output signal of a plate-like capacitive wetness element, proving adequate agreement between modelling and measurement. Fig. 1b gives an indication for the quality of the calculation schemes for dew and rain induced leaf-wetness periods on top of an apple orchard and of a vineyard (HOPPMANN 1996).

3. USE OF THE RADAR NETWORK FOR LEAF- WETNESS ESTIMATIONS

Input data for the leaf-wetness calculation schemes are usually taken from the wide-meshed synoptic network. In order to obtain a higher spatial resolution, experiments are currently being carried out to use the hourly radar measurements of rainfall. Fig. 2 shows an image of rain duration for a vine-growing area in Rheinhessen. The image consists of 51*51 grid cells, each one 1km*1km in dimension.
 

Fig.2: Rain duration derived from the weather radar output (Rheinhessen, 20 May 1997, 06-18 UTC).
 

4. APPLICATION OF LEAF-WETNESS DURATION
 

In a number of subprogrammes the wetness-dependent behaviour of fungal plant diseases is calculated. Fig. 3a shows the disease potential of the Plasmopara viticola pathogen in vineyards. Only the leaf-wetness duration between 6 a.m. and 6 p.m. is regarded because infections can start only during darkness.

Fig.3a: Comparison of Plasmopara viticola model runs supplied with sensor-measured and with simulated leaf-wetness data at Geisenheim for the period May - July 1995

The minimum wetness duration for a beginning infection is 4 hours while the air temperature has to exceed 11 ^oC during the wetness event. Infections become visible on the leaves after the incubation period of 5 - 10 days depending on the temperature. For example, the steep increase in the step-like infection curve occuring on 16 July may be caused by the nocturnal rain period of the 11th of July. It is assumed that economic losses do not occur as long as the number of oilspots on leaves remains below a threshold of 5000 spots per hectare. On 8 July this critical level is exceeded, so that spray applications can be recommended.

Fig.3b: Effect of leaf-wetness duration on apple-scap infection periods indicated by an infection index greater than 127 (Braunschweig, May 1997).
 

Another example is the occurrence of the apple scab disease caused by the fungus Venturia inaequalis in apple orchards (Arauz et al 1990).
 

In spring ascospore infection periods start only during daylight conditions after a temperature dependent minimum wetness-duration of at least eight hours (Fig. 3b). If the infection index reaches a critical level of 127 a visible light infection occurs after an incubation period of 8 - 15 days as long as the fungus is not counteracted by fungicides in order to avoid crop yield losses. The index value of 600 on day 141 indicates a severe infection.

5. DATA SUPPLY AND DATA LINKS TO FARMERS AND VINEGROWERS

Fig.4: Data flow from the networks to the end user

Microclimatic models are used for leaf-wetness estimations. These models are supplied with data from the DWD's standard weather observation network, from the weather forecast model and, tentatively, from the weather radar.

After linking crop models to the leaf-wetness schemes the Deutscher Wetterdienst through its agrometeorological advisory service supports the farmers' control and management programmes. All the crop relevant information is made available by the DWD in a time-critical way through fax and local telephone messages, and by modem links to individual users via existing computer communication networks (online services). In future it will be possible to distribute information via internet providers.
 

Literature

Arauz, L. F., T. B. Sutton, L. R. Pope: Simultaneous use of infection criteria for three apple diseases for timing of fungicide sprays. – Phytopathology 80, 1212–1218, 1990.

Barthakur, N. N.: A comparative study of radiometric and electronic leaf wetness sensors. – Agric. For. Meteorol. 36, 83–90, 1985.

BRADEN,H.: The model AMBETI: a detailed description of a soil-plant-atmosphere model. - Ber. des Deutschen Wetterdienstes 195, VI, 117 S, 1995

Hoppmann, D.: Agrarmeteorologische Entscheidungsmodelle im Weinbau am Beispiel der Prognose der Rebenperonospora (Plasmopara viticola). – Abschlußbericht AM 37, Deutscher Wetterdienst, Geisenheim, 75 p., 1996.

Hoppmann, D., WITTICH, K.P.:Epidemiology-related modelling of the leaf-wetness duration as an alternative to measurements, taking (Plasmopara viticola) as an example – J. of Plant Dis. and Prot. 104, 533-544, 1997

Wittich, K.-P.: Ansätze zur Abschätzung der Blattbenetzungsdauer. – DWD intern 54 (Beiträge zur Agrarmeteorologie 1/93), Deutscher Wetterdienst, Offenbach, 44 p., 1993.

Schrödter, H., J. Ullrich: Weitere Untersuchungen zur Biometeorologie und Epidemiologie von Phytophthora infestans (Mont.) de By. Ein neues Konzept zur Lösung des Problems der epidemiologischen Prognose. – Phytopath. Zeitschr. 56, 265–278, 1966.

Werres, G., H. Hindorf: Development of a forecasting model for scald, Rhynchosporium secalis, on rye in relation to climate conditions. – EPPO Bulletin 21, 485–493, 1991.