Leaf Wetness Duration Modeling for Plant Disease Management

Leaf wetness duration (LWD) is a principal driver of many foliar diseases, acting as the window during which pathogenic spores germinate and infect plant tissues. Accurate estimation of LWD underpins decision support systems used to time fungicide applications, reduce chemical inputs and mitigate crop losses. Modelling approaches range from mechanistic simulations of dew formation and evaporation based on energy balance and microclimatic variables, to empirical and machine-learning models calibrated against sensor observations. Advances in high-resolution remote sensing, image analysis and artificial intelligence have improved spatial and temporal coverage of LWD predictions, expanding their applicability from research stations to commercial farms and advising platforms. Recent efforts have emphasised the integration of satellite observations, on-leaf imaging and networked sensor data to account for canopy structure, leaf orientation and microclimatic heterogeneity. By linking LWD models to epidemic risk algorithms, growers can adopt site-specific disease management, tailoring fungicide schedules to local weather patterns and canopy conditions. This integrative approach has global significance, offering sustainable strategies for diverse cropping systems from protected greenhouse production to open-field horticulture.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Leaf Wetness Duration Modeling for Plant Disease Management publication trend

The graph below shows the total number of articles in leaf wetness duration modeling for plant disease management across all publications each year (not limited to Nature Index journals).

Technical terms

Leaf wetness duration (LWD): The cumulative time that water remains on the leaf surface, critical for pathogen infection and disease development.

Generalised additive model (GAM): A flexible statistical model that relates predictors to a response through smooth functions, allowing non-linear relationships.

Convolutional neural network (CNN): A class of deep learning model structured to process grid-like data such as images, extracting hierarchical features for classification or regression tasks.

Machine learning (ML): A collection of algorithms that learn patterns from data to make predictions or decisions without explicit programming.

Geostationary satellite: A satellite in an equatorial orbit matching Earth’s rotation, providing continuous observation of the same geographical area for meteorological monitoring.

References

1. Non-contact leaf wetness measurement with laser-induced light reflection and RGB imaging. Biosystems Engineering (2024).
2. Utilizing High-Resolution Imaging and Artificial Intelligence for Accurate Leaf Wetness Detection for the Strawberry Advisory System (SAS). Sensors (2024).
3. Prediction of Leaf Wetness Duration Using Geostationary Satellite Observations and Machine Learning Algorithms. Remote Sensing (2020).