Abstract:The objective of this work was to determine models for the estimation of leaf wetness percentage at three heights in the soybean (Glycine max) canopy, using meteorological variables from stations installed at the crop site and at an agrometeorological station. The experiment was conducted in three harvest seasons, in an area cropped with soybean, in the municipality of Londrina, in the state of Paraná, Brazil. To collect the meteorological variables, electronic trees were installed at four heights (0.3, 0.6, 0… Show more
“…Finally, source code and result logs are available at https://git.code.tecnalia.com/afarcloud/dss algorithms/leaves wetness. 3…”
Section: Methodsmentioning
confidence: 99%
“…In this context, the lack of a standard method for calculating and measuring the Leaf Wetness Duration (LWD) is arguably the most concerning constraint for achieving rigorous data [3]. This causes many practical difficulties when comparing results, and may lead to bad decisions made by growers about the diseases related to the leaf wetness.…”
Leaf wetness often emerges as the result of the exchange of atmospheric water-soluble gases between the Earth surface and the atmosphere. The importance of this feature resides in the relationship that exists between leaf wetness and various plant diseases. In order to measure this variable, there is a need for deploying physical sensors to capture wetness readings of a crop area. However, the installation and maintenance of these sensors is a hard task that involves qualified people, time and high economical costs. Moreover, the acquisition, storage and analysis of data must be taken into consideration to infer this information and issue countermeasures preemptively. This work presents a leaf wetness soft-sensing approach that relies on predictive machine learning models to estimate the wetness of the leaves of a specific crop. Specifically, among the learning algorithms that are evaluated for this purpose, we include Random Vector Functional Link (RVFL) networks, a family of neural networks that embrace randomization at their core to yield a highly efficient training process. By virtue of machine learning, physical sensors can be replaced by soft-sensors capable of providing the information related to the wetness of the leaves of the crop. In this way, human effort and costs are largely reduced, while ensuring a high precision of the wetness estimation as proven by experiments with real-world data.
“…Finally, source code and result logs are available at https://git.code.tecnalia.com/afarcloud/dss algorithms/leaves wetness. 3…”
Section: Methodsmentioning
confidence: 99%
“…In this context, the lack of a standard method for calculating and measuring the Leaf Wetness Duration (LWD) is arguably the most concerning constraint for achieving rigorous data [3]. This causes many practical difficulties when comparing results, and may lead to bad decisions made by growers about the diseases related to the leaf wetness.…”
Leaf wetness often emerges as the result of the exchange of atmospheric water-soluble gases between the Earth surface and the atmosphere. The importance of this feature resides in the relationship that exists between leaf wetness and various plant diseases. In order to measure this variable, there is a need for deploying physical sensors to capture wetness readings of a crop area. However, the installation and maintenance of these sensors is a hard task that involves qualified people, time and high economical costs. Moreover, the acquisition, storage and analysis of data must be taken into consideration to infer this information and issue countermeasures preemptively. This work presents a leaf wetness soft-sensing approach that relies on predictive machine learning models to estimate the wetness of the leaves of a specific crop. Specifically, among the learning algorithms that are evaluated for this purpose, we include Random Vector Functional Link (RVFL) networks, a family of neural networks that embrace randomization at their core to yield a highly efficient training process. By virtue of machine learning, physical sensors can be replaced by soft-sensors capable of providing the information related to the wetness of the leaves of the crop. In this way, human effort and costs are largely reduced, while ensuring a high precision of the wetness estimation as proven by experiments with real-world data.
“…Research on LWD has mostly focused on crops and grasslands, and the monitoring position of leaf wetness has been limited to the top of canopy (Sentelhas et al, 2008;Park et al, 2019;Wang et al, 2019). Only a few studies monitored LWD at different canopy positions (Igarashi et al, 2018). In fact, due to complex canopy structure and variable canopy micro-meteorology, canopy wetness is variable in both time and space (Bassimba et al, 2017;Binks et al, 2021).…”
Leaf wetness provides a wide range of benefits not only to leaves, but also to ecosystems and communities. It regulates canopy eco-hydrological processes and drives spatial differences in hydrological flux. In spite of these functions, little remains known about the spatial distribution of leaf wetness under different soil water conditions. Leaf wetness measurements at the top (180 cm), middle (135 cm), and bottom (85 cm) of the canopy positions of rainfed jujube (Ziziphus jujuba Mill.) in the Chinese loess hilly region were obtained along with meteorological and soil water conditions during the growing seasons in 2019 and 2020. Under soil water non-deficit condition, the frequency of occurrence of leaf wetness was 5.45% higher at the top than at the middle and bottom of the canopy positions. The frequency of occurrence of leaf wetness at the top, middle and bottom of the canopy positions was over 80% at 17:00-18:00 (LST). However, the occurrence of leaf wetness at the top was earlier than those at the middle and bottom of the canopy positions. Correspondingly, leaf drying at the top was also latter than those at the middle and bottom of the canopy positions. Leaf wetness duration at the middle was similar to that at the bottom of the canopy position, but about 1.46-3.01 h less than that at the top. Under soil water deficit condition, the frequency of occurrence of leaf wetness (4.92%-45.45%) followed the order of top>middle>bottom of the canopy position. As the onset of leaf wetness was delayed, the onset of wet leaf drying was advanced and the leaf wetness duration was shortened. Leaf wetness duration at the top was linearly related (R 2 >0.70) to those at the middle and bottom of the canopy positions under different soil water conditions. In conclusion, the hydrological processes at canopy surfaces of rainfed jujube depended on the position of leaves, thus adjusting canopy structure to redistribute hydrological process is a way to meet the water need of jujube.
“…(b) There are different types of sensors, and the measurements vary according to type [7]; thus, two sensors of different types in the same place could generate different results. (c) There is no standard protocol for installation of sensors and measurement of LWD, and even the specifications change according to sensor brand [12]. Sensor used by ICAFE and sensor from another brand.…”
The prediction of leaf wetness duration (LWD) is an issue of interest for disease prevention in coffee plantations, forests, and other crops. This study analyzed different LWD prediction approaches using machine learning and meteorological and temporal variables as the models’ input. The information was collected through meteorological stations placed in coffee plantations in six different regions of Costa Rica, and the leaf wetness duration was measured by sensors installed in the same regions. The best prediction models had a mean absolute error of around 60 min per day. Our results demonstrate that for LWD modeling, it is not convenient to aggregate records at a daily level. The model performance was better when the records were collected at intervals of 15 min instead of 30 min.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.