Ultra-high-resolution UAV-imaging and supervised deep learning for accurate detection of Alternaria solani in potato fields.

Alternaria solani is the second most devastating foliar pathogen of potato crops worldwide, causing premature defoliation of the plants. This disease is currently prevented through the regular application of detrimental crop protection products and is guided by early warnings based on weather predictions and visual observations by farmers. To reduce the use of crop protection products, without additional production losses, it would be beneficial to be able to automatically detect Alternaria solani in potato fields. In recent years, the potential of deep learning in precision agriculture is receiving increasing research attention. Convolutional Neural Networks (CNNs) are currently the state of the art, but also come with challenges, especially regarding in-field robustness. This stems from the fact that they are often trained on datasets that are limited in size or have been recorded in controlled environments, not necessarily representative of real-world settings. We collected a dataset consisting of ultra-high-resolution modified RGB UAV-imagery of both symptomatic and non-symptomatic potato crops in the field during various years and disease stages to cover the great variability in agricultural data. We developed a convolutional neural network to perform in-field detection of Alternaria, defined as a binary classification problem. Our model achieves a similar accuracy as several state-of-the-art models for disease detection, but has a much lower inference time, which enhances its practical applicability. By using training data of three consecutive growing seasons (2019, 2020 and 2021) and test data of an independent fourth year (2022), an F1 score of 0.93 is achieved. Furthermore, we evaluate how different properties of the dataset such as its size and class imbalance impact the obtained accuracy.

Perspectives for Remote Sensing with Unmanned Aerial Vehicles in Precision Agriculture.

Remote sensing with unmanned aerial vehicles (UAVs) is a game-changer in precision agriculture. It offers unprecedented spectral, spatial, and temporal resolution, but can also provide detailed vegetation height data and multiangular observations. In this article, we review the progress of remote sensing with UAVs in drought stress, in weed and pathogen detection, in nutrient status and growth vigor assessment, and in yield prediction. To transfer this knowledge to everyday practice of precision agriculture, future research should focus on exploiting the complementarity of hyperspectral or multispectral data with thermal data, on integrating observations into robust transfer or growth models rather than linear regression models, and on combining UAV products with other spatially explicit information.

Capacitive water release and internal leaf water relocation delay drought-induced cavitation in African Maesopsis eminii.

The impact of drought on the hydraulic functioning of important African tree species, like Maesopsis eminii Engl., is poorly understood. To map the hydraulic response to drought-induced cavitation, sole reliance on the water potential at which 50% loss of xylem hydraulic conductivity (ψ50) occurs might be limiting and at times misleading as the value alone does not give a comprehensive overview of strategies evoked by M. eminii to cope with drought. This article therefore uses a methodological framework to study the different aspects of drought-induced cavitation and water relations in M. eminii. Hydraulic functioning of whole-branch segments was investigated during bench-top dehydration. Cumulative acoustic emissions and continuous weight measurements were used to quantify M. eminii's vulnerability to drought-induced cavitation and hydraulic capacitance. Wood structural traits, including wood density, vessel area, diameter and wall thickness, vessel grouping index, solitary vessel index and vessel wall reinforcement, were used to underpin observed physiological responses. On average, M. eminii's ψ50 (±SE) was -1.9 ± 0.1 MPa, portraying its xylem as drought vulnerable, just as one would expect for a common tropical pioneer. However, M. eminii additionally employed an interesting desiccation delay strategy, fuelled by internal relocation of leaf water, hydraulic capacitance and the presence of parenchyma around the xylem vessels. Our findings suggest that exclusive dependence on ψ50 would have misdirected our assessments of M. eminii's drought stress vulnerability. Hydraulic capacitance linked to anatomy and leaf-water relocation behaviour was equally important to better understand M. eminii's drought survival strategies. Because our study was conducted on branches of 3-year-old greenhouse-grown M. eminii seedlings, the findings cannot be simply extrapolated to adult M. eminii trees or their mature wood, because structural and physiological plant properties change with age. The techniques and methodological framework used in this study are, however, transferable to other species regardless of age.

Early detection of Psa infection in kiwifruit by means of infrared thermography at leaf and orchard scale.

Pseudomonas syringae pv. actinidiae (Psa), the causal agent of bacterial canker of kiwifruit, has become a worldwide threat for the kiwifruit industry. In this work, the potential of infrared thermography for early detection of physiological symptoms related to Psa-infection at leaf and at orchard block scale was assessed. At the leaf level, thermal cold spots appeared shortly after Psa-infection, well before any visual symptoms. A few weeks after infection, thermal hot spots were observed, associated with, but not limited to, spots of visible leaf necrosis. At orchard block level, Psa-infected canes were significantly warmer in both blocks and on all measurement days. A novel wet reference surface, existing of a cluster of cotton imitation leaves with similar dimensions and orientation as real leaves and remaining wet through sucking up water from a small container, was used to estimate the crop water stress index (CWSI). CWSI showed stable values of infected and uninfected areas during the day and between following days. Crop temperature and CWSI were closely correlated with leaf stomatal conductance, which was lower in infected canes. A Psa-infection map based on canopy temperature revealed that Psa infects the outer canes rather than the central part of the canopy.

Assessment of land use impact on water-related ecosystem services capturing the integrated terrestrial-aquatic system.

Although the importance of green (evaporative) water flows in delivering ecosystem services has been recognized, most operational impact assessment methods still focus only on blue water flows. In this paper, we present a new model to evaluate the effect of land use occupation and transformation on water quantity. Conceptually based on the supply of ecosystem services by terrestrial and aquatic ecosystems, the model is developed for, but not limited to, land use impact assessment in life cycle assessment (LCA) and requires a minimum amount of input data. Impact is minimal when evapotranspiration is equal to that of the potential natural vegetation, and maximal when evapotranspiration is zero or when it exceeds a threshold value derived from the concept of environmental water requirement. Three refinements to the model, requiring more input data, are proposed. The first refinement considers a minimal impact over a certain range based on the boundary evapotranspiration of the potential natural vegetation. In the second refinement the effects of evaporation and transpiration are accounted for separately, and in the third refinement a more correct estimate of evaporation from a fully sealed surface is incorporated. The simplicity and user friendliness of the proposed impact assessment method are illustrated with two examples.