The wine industry must face many challenges because of climate change. One of them is the increase in temperatures and droughts events. These changes sometimes lead to yield losses and can also impact the quality of the wine produced (e.g., increased alcohol content). The management of available water is also a sensitive issue as water requirements for vineyard irrigation are quickly increasing in the south of France. In this context, there is a need for a decision tool that can help evaluate the vine water status through the entire growth season at a large scale. To address this issue, we have previously developed a model (see Laroche-Pinel et al. 2021a) which predicts the vine Stem Water Potential (Ψstem) using Sentinel-2 (S2) images. This model was developed based on a field campaign over three years. The present study now aims to investigate the feedback of winegrowers on the outputs of our model. Therefore, it was applied on the plots of five wine estates that do not belong to the set used in the initial paper. The qualitative results show interesting spatial and temporal consistency in accordance with winegrower knowledge, irrigation data, and weather forecast. The predicted Ψstemhighlights spatial variability in vine fields where a water source emerges and reflects the differences between vine fields with a drip or sparkling irrigation or without an irrigation system. The predicted Ψstem also clearly reacts to a peak in temperature. According to their feedback, three of the five winegrowers would be glad to use this service in the years to come.
Viticulture is highly affected by climatic conditions, as much for its yield as for the quality of the grapes and therefore the wine produced (Chaves et al., 2007; Costa et al., 2007). According to van Leeuwen and Darriet (2016), this is the main factor affecting plant development and berry composition ahead of the soil or grape variety. The climate of the Mediterranean region was until the end of the 1990s rather favourable to the cultivation of vines with wet winters and warm summers, and balanced rainfall. Since 2000, the scarcity of rainfalls and the increase in evapotranspiration lead to regular and important water deficits during the growing period of vineyards (April–June), and thus the need for irrigation to maintain proper yield and wine quality. This raises the question of water resource management (Parker et al., 2020). At the same time, thoughts are ongoing to find other solutions to water stress based on alternative agronomic practices (soil and inter-row management, vine pruning), variety choice, or plant density for example (Bernardo et al., 2018). To implement these different solutions, it is necessary to first identify and quantify the needs. The first need is to detect the areas that suffer the most from water stress to optimise water use at the territory level. The second is the monitoring of the water status throughout the season to adapt irrigation management or other agricultural practices at the field level and maintain harvest and quality targets. A common, robust, and precise method to measure vine water status is the Ψstem (Acevedo-Opazo et al., 2008). Thanks to pressure chamber measurements (Scholander et al., 1965), Ψstem reflects plant transpiration and soil conductivity. It reflects the capacity of the vine to conduct water from the soil to the atmosphere (Choné et al., 2001). However, this method is time-consuming and cannot be applied at a large scale during all the growth season.
Remote sensing could be a relevant tool as it gives spatial information on vegetation with possibly an interesting temporal resolution. Optical images have already been used in several studies about vine water status for irrigation management with UAV multispectral images (Romero et al., 2018), to estimate Ψstem with Landsat 8 images (Borgogno-Mondino et al., 2018) or with UAV multispectral images (Poblete et al., 2017) and to identify irrigation zones with an aircraft (Bellvert et al., 2012).
Considering S2, its spatial resolution of 10 to 20 m for most of its 13 bands, and its temporal resolution of 5 days make this satellites constellation especially suited for vegetation monitoring (Addabbo et al., 2016; Weiss et al., 2020). Despite the inability of S2 to differentiate the row from the inter-row on the images (Cogato et al., 2019; Di Gennaro et al., 2016; Sozzi et al., 2020), the combination of its spatial, spectral, and temporal resolution makes it a very interesting tool to monitor vineyards at a large scale (Devaux et al., 2019) taking into account spatial variability (Di Gennaro et al., 2019). Moreover, its potential use for vineyard monitoring has already been demonstrated to identify vine water status by Cohen et al. (2019) and to quantify the impact of heatwaves on the irrigated vineyard by Cogato et al. (2019). The ultimate advantage of S2 for future operational service is that it could be low cost as the access to the acquired images is free and open.
In a previous study, we demonstrated the effectiveness of a regression model to predict vine Ψstem using S2 images (Laroche-Pinel et al., 2021a). The study proposed here aims to check the consistency of the results obtained with the model in a pre-operational context for several new vine fields throughout the 2020 season. The goal is to verify spatial and temporal consistency of the results in a qualitative way based on winegrowers' knowledge and expertise and also using weather forecasts and in-field information. These exchanges with potential end-users also constitute an interesting exploratory step to assess the interest of a potential new decision support tool.
Section “Material and method” presents the characteristics of the model and how it has been applied in this study, including the details of the five wine estates’ vine fields, the S2 images processing, how Ψstem are usually interpreted and how the exploitation of the results is made based on winegrowers experience. Section “Results” describes the exploration of the spatial and temporal variability, with few representative results. Section “Discussion” first focus on the consistency of the results according to winegrowers feedbacks, then reports the limitations of the model and discuss how this model could be used in an operational service. Section “Conclusion” resume the goal of the project, the main results, and the future experiments to come.
Material and method
1. Model development
The model previously developed and described in Laroche-Pinel et al. (2021a) was based on knowledge previously obtained using hyperspectral measurements (Laroche-Pinel et al., 2021b) and was calibrated using in-field Ψstem measurements. More than 2500 measurements were obtained using a pressure chamber over 3 years (2018, 2019, 2020) on 36 vines fields in the Mediterranean region. In this study, five supervised regression machine learning algorithms were tested to find possible relationships between stem water potential and Sentinel-2 images (bands reflectance values and vegetation indices). The best regression model allows one to predict Ψstem using Red (B4), NIR (B8), Red-Edge (B6) and SWIR (B11) bands reflectance of S2 images (R² = 0.40, RMSE = 0.26).
2. Application of the model
For this new study, the model was applied to cloud-free images over 170 new vineyard fields, considering only the pixels fully included in the field. The model processes one image at a time. The results obtained are maps at the pixel level (intra-field variability) and the field level (average of all the pixel values included in the plots). Predicted Ψstem values are also extracted for each field at each date for temporal monitoring. The goal of the study is to ensure the consistency of this prediction using winegrowers' expertise, weather and water supply data. Figure 1 below shows all the different steps.
Figure 1. Flowchart of the method used in this study (S2: Sentinel-2, Ψstem, Water Potential).
2.1. Study sites
In this study, we worked with five wine estates in the south of France whose winegrowers well know their fields and closely monitor their water status. In total, 170 vine fields were monitored near Carcassonne and Beziers (Figure 2). This region benefits from a Mediterranean climate with hot, dry summers and cool, wet winters. Vine fields are mostly managed by Cordon, with 2 m inter-row, bare soil and cover 220 ha in total. The database was diversified with 18 grape varieties, 11 red (Syrah, Alicante, Cabernet-Sauvignon, Carignan, Cinsault, Grenache, Mourvedre, Marselan, Petit verdot, Pinot noir, Merlot) and 6 white (Chardonnay, Roussanne, Marsanne, Vermentino, Muscat, Viognier), different types of irrigation (no irrigation, soil and aerial system), and soils (clay or limestone). The planting dates range from 2002 to 2015.
2.2. Sentinel -2 images processing
The S2 sensor takes images with a swath of 290 km and images are then cut into sections of 110 km × 110 km named “tiles”. The studied area is covered by three tiles as shown in Figure 2 (T31TDH, T31TDJ and T31TEJ). All available S2 images from June to September 2020 were downloaded from the THEIA platform in L2A format with a cloud mask and an atmospheric correction (CNES, 2021). An average of 20 images could be processed for each field, this means that we had a usable image (without cloud) every 7 days on average (approximately 10 images for each field from June to September). Among the 13 bands of S2, the model developed by Laroche-Pinel et al., 2021a estimates Ψstem from the 4 bands in Red, Red-Edge, NIR and SWIR spectral domains with a spatial resolution of 10 or 20 m.
The 20 m bands are resampled to 10 m and then, the first step of the process consisted in applying a buffer to avoid border effects as done by Devaux et al. (2019). A buffer of 5 m is applied inside the field boundary and only the pixels entirely included in the boundaries are kept. In the second step, the model is applied to each image's pixels to have a map of the predicted Ψstem by pixel for the intra field variability, then values are averaged by field to highlight the inter-field variability. The last step consists of extracting Ψstem estimation by field for each date for temporal monitoring. These steps are summarised and illustrated in Figure 3.
Figure 3. Illustration of the process for one field (S2: Sentinel-2). The first step consisted in applying a buffer inside the original field boundary. Then the model is applied in each pixel and Ψstem predicted values are averaged by field. The mean value of each field is finally extracted by date for temporal monitoring.
2.3. Ψstem interpretation
The model predicts vine water status by predicting Ψstem from 0 MPa (no water constraint) to -2 MPa (high water constraint). Water stress can occur more or less quickly, depending on climatic conditions (van Leeuwen et al., 2009) and the response speed of the Ψstem will also differ depending on it. Ψstem values can be interpreted in terms of water constraint according to the vine development stage and the desired future wine (Payan, 2007; Salva, 2016). Indeed, the required water constraint will differ from a lower constraint at pea-size (to ensure the correct development of the grape) to a higher constraint at the ripening stage (to enhance grape quality by concentrating aromatic compounds). Additionally, for ageing red wine, a high water constraint will be required unlike for a light white wine for example. In all cases, a Ψstem value below –1.6 MPa is indicative of excessive water stress for vines, which may cause irreversible damage for future production (Deloire et al., 2020).
2.4. Exploration of model application results with winegrowers interviews
At the beginning of the study, vine field boundaries with information about the grape variety and irrigation management were collected from winegrowers. Additional useful information were also collected: weather record with daily mean temperature and rainfalls from June to September, and the irrigation programs reporting dates and quantities of water brought by drip irrigation. Our Ψstem prediction model has been processed at the end of the summer on all available and usable images. Finally, a meeting with the winegrowers has been organised on November 26th, 2020, to present our work and show them the maps and the temporal monitoring of their vine fields. They had to give feedback and discuss the consistency of the model.
The first thing to be evaluated is the spatial consistency of the predicted Ψstem maps over wine estates fields i.e., to assess if the observed spatial patterns are correlated to the actual vine water status. Maps highlighting intra- or inter-field variability over wine estates were shown to each winegrower to check the model consistency according to their field knowledge (e.g., different soil composition, missing vines, irrigation).
Secondly, to evaluate the temporal consistency of the predicted water status over time, Ψstem temporal monitoring of vine fields were produced for each field and presented to winegrowers The Ψstem estimation evolution was then analysed with regards to weather data and irrigation management. As described previously, vine water status evolves from pea-size to harvest. If the decrease is greater or lesser than expected or if a big step is observed between close dates, it could either reflect an alarming change in water status or other agronomical practices (e.g., cover grass destruction, canopy trimming or tying-up).
1. Highlighting Spatial variability
The goal here is to verify that the model gives correct estimation by analysing the predicted Ψstem variations at the intra-field scale. The following figures show three examples of maps with the feedbacks from winegrowers.
The first example in Figure 4 shows the predicted Ψstem map on August the 10th for several red (red border) and white (yellow border) vineyards of a wine estate. The colour scale of the map ranges from dark blue for no water constraint (0 MPa) to dark orange for severe water stress (–2 MPa).
Looking at this map, the winegrower explained that there is a water source close to the areas that stand out in blue which may explain why there is less stress in this location. Moreover, he said that the field located in the northwest has a different soil composition in the lower-left corner that does not fix water as well as the rest of the field, and can be seen with a greener colour.
Figure 4. Predicted water status map on August 10th, 2020.
The second example in Figure 5 is a predicted water status map on July the 14th. Concerning the white varieties, two fields (1 and 2) of the same variety (Chardonnay) seem to have different water statuses. The second one seems to be less stressed. Moreover, the fields in the east stand out as being less stressed than those from the west. Finally, it can also be noticed that red fields seem to be less stressed than the white ones.
Thanks to the discussion with the winegrower, we learned that fields 1 and 2 have different planting years (field 1: 2012 and field 2: 2003). This makes sense as the older vine, with a deep adult root system, seems to have a better water status than the youngest, with a young and still shallow root system. Moreover, he also described the east area as wetter which can explain the difference between the fields in the east and west. The difference between red and white varieties also makes sense since white varieties do not stand too much water constraint, as outlined in the section “Materials and Method. 2.3.”.
Figure 5. Predicted water status map on July 14th, 2020 (field 1: chardonnay of 8 years old, field 2: chardonnay of 17 years old).
The third example shown with the map in Figure 6 highlights two fields in the middle that seem to have a more stressed area in their northern parts.
According to the winegrower, these areas have many missing vines (a rate higher than 20 %). Such an abnormal level of missing vines may affect the model of Ψstem estimation by increasing the soil fraction seen by the satellite.
2. Monitoring temporal evolution
The second objective is to verify the relevance of the temporal monitoring. It is important to know whether the model reacts correctly throughout the season and if the variations observed are actually due to changes in the water status of the vines. Four examples are detailed below to illustrate how the relevance of the model monitoring was verified.
2.1. Link with weather data
Figure 7 emphasises the link between the mean temperature recorded in a wine estate during summer 2020 and the predicted Ψstem of its 32 vine fields (mean and standard deviation). Indeed, Ψstem evolution is inversely proportional to the mean temperature which means that the model reacts in the right way throughout the season. According to the intensity and the duration of the temperature increase, the answer of the vine to a temperature increase is more or less quick, but as shown with the red boxes in Figure 7, the Ψstem estimation is always impacted by high or long temperature increases. For example, in the first box, it can be clearly seen that the predicted Ψstem is inversely proportional to the temperature increase. For the others, the correlation depends on the intensity and duration of the temperature increase. In the third box, the predicted Ψstem considerably decreases after several days of high temperatures.
2.2. Evaluation of the Ψstem prediction over a whole wine estate after a rain and during a heatwave
Figure 8 shows two water status maps at a field scale for two wine estates on July the 26th and July the 31st. Figure 8a,c displays water status three days after rain (66.4 mm on July the 23rd) and Figure 8b,d highlights the water status during a temperature peak (mean temperature between 25 °C and 27 °C). As expected, the majority of the fields tend to be more stressed in Figure 8b,d but their levels remain reasonable for the first wine estate (Figure 8b). The winegrower explained to us that it is not surprising from his point of view because he tries to irrigate all his vineyard as soon as they require it with a drip irrigation system. On the contrary, for the second estate (Figure 8d), the colour of the plots tend to be more orange for some fields indicating that there is a big decrease in predicted Ψstem values. The winegrower of this second estate explained to us that it is logical as he cannot irrigate all these fields in the same way, and some are never irrigated.
Figure 8. Water status maps averaged by field for two wine estates at two dates, a) and c) after the rain on July 26th, 2020 and b) and d) during a temperature peak on July 31st, 2020. Figures a) and b) represent a well-irrigated wine estate and Figures c) and d) represent a wine estate with non-irrigated plots.
2.3. Impact of irrigation management
To follow and compare different vine fields with an interesting range of water statuses, it was particularly interesting to have access to a wine estate having a different irrigation system installed on their fields: 15 vine fields have no irrigation system, 15 benefit from an overhead sprinkler irrigation system and 19 have a drip irrigation system. Figure 9 shows the temporal monitoring of predicted stem water potential for these fields according to their irrigation systems. As expected, fields without irrigation systems tend to be more stressed than fields with soil or aerial systems. Fields with a drip irrigation system tend to be less stressed than the others: drip irrigation allows a better spatial and temporal regularity of irrigation, with fewer losses than sparkling irrigation. Furthermore, the aerial irrigation system was broken in early July. The fields may then not have received the right amount of water and this is visible on the predicted Ψstem value as a rapid decrease is observed at these dates for fields with aerial irrigation system (red box in Figure 9).
2.4. Verification of watercourse according to the objective of an ageing red wine
In Figure 10, Ψstem thresholds were reported on temporal monitoring according to a watercourse adapted for an ageing red wine (Ojeda and Saurin, 2014). Proper management of the whole estate is highlighted as predicted Ψstem are almost always in the optimal range, i.e., in “Balanced water status”. This estate is the same as the one shown in Figure 8 and the winegrower well manages the water status of its fields by having a global vision of its entire estate and by being able to irrigate each of its fields when they need it with a drip system. In this figure, the temperature peak at the end of July can also be noticed with a decrease of the predicted Ψstem at the end of July.
1. Winegrowers feedback
Among the five solicited winegrowers, three were enthusiastic about the model’s results on their vine fields. Among these three, two would even like to use a service based on this model to help them better choose the fields to irrigate and the third one would rather rely on both the Ψstem prediction and a vigour estimation to improve the management of their fields (pruning, harvest date and allotments). Another winegrower would like to test the model for several years to check the consistency of the results over several vintages, and the last one considers that he knows his plots well enough to avoid using this kind of service. In fact, his wine estate is smaller than the others (less than 50 ha) and he can visit these fields very regularly during the season.
1. 1. Consistency of spatial variabilities
The feedback from winegrowers was positive with the right correlation between the predicted water status at an intra-field level and the knowledge provided by the winegrower. Moreover, differences between red and white cultivars have also been highlighted which seems logical as they do not require the same water constraints, and are planted in deeper soils, or more irrigated (van Leeuwen et al., 2009). Our Ψstem predictions have also highlighted that older vine fields seem to be less stressed than young and this makes sense as they are generally less sensitive to water deficit because of their deeper root system (Bou Nader et al., 2019).
1.2. Consistency of temporal evolutions
The first positive aspect of the temporal monitoring with our model is the good relationship with the weather in relation to heatwaves (Figure 7) and rain (Figure 8) events. The second is the distinction between irrigated and non-irrigated vine fields of an entire vine estate (Figure 9), this means that the model can reasonably be used to distinguish fields with a lack of water. This also proves the robustness of the model over a whole wine estate even with several water supply practices. Finally, as the future service will provide Ψstem estimation during the season in real-time, it could also be possible to verify and adjust the water management of a vine estate with water status thresholds according to the development stage and desired future wine (Figure 10).
2. Limitations of the model
To take into account the risk of water stress overestimation due to a high missing rate or a low vine vigour, it is necessary to measure the vine vigour as a complement with vegetation indices or biophysical parameters, for example. A threshold could be determined below which the water stress level would not be calculated because there would not be enough vegetation. Moreover, to avoid underestimating stress, the vegetation in the inter-row can be useful information to better interpret the Ψstem prediction. In fact, in Mediterranean region, the grass is not very vigorous in summer, so in our studies, the presence of grass between the vine rows was not a problem.
This model is based on the prediction of Ψstem which is a commonly used precise vine water status assessment, however, recent studies have highlighted some limits of this measurement under certain climatic conditions (Santesteban et al., 2019, Suter et al., 2019) or for isohydric behaviour (Blanco-Cipollone et al., 2017).
New field campaigns are needed to go further with the exploration and validation of the model on vine fields in other regions with different varieties, hydric behaviour, inter-row management or vine pruning, for example, that could lead to signal differences or/and could impact the vineyard signature in 10 or 20 m pixels.
A way to enhance the precision of the model could be to make some Ψstem calibration measurements each year at strategic locations to adjust the prediction of the model. More information like rain or average temperatures could also be used as additional predictors in the model.
Finally, the model used in this study combines four bands of an optical multispectral satellite (S2) while the arrival and development of thermal, radar or hyperspectral satellites could be interesting to explore to improve the accuracy of the vine water status prediction models as they allow one to assess other characteristic links to vegetation water status (e.g., temperature, complete vegetation spectrum) (Das et al., 2020, Pagay & Kidman, 2019) or soil moisture (El Hajj et al., 2018, Balenzano et al., 2021). However, most of these tools are not yet available to use from an operational point of view at an affordable price for customers.
3. Towards an operational service
To operate a future service, two levels seem to emerge, one to help winegrowers to manage their vine fields at a vine estate scale and one for winegrowers unions to have an entire view of a region for example. In both cases, spatial and temporal information will be needed. Yearly temporal monitoring will help to know where to start watering for example. Maps can help in designing irrigation plans, with irrigation blocks corresponding to areas with the same water requirements. Pluriannual monitoring would give a better idea of the characteristics of each vintage, terroir, and region. Both can contribute to justifying applications for irrigation authorisation adapted to each area and year climatic characteristics, and also to long term policies concerning water and vineyard management. Finally, it would be interesting to add the weather information (historical and forecast) to have an overview of all available data. This would allow to complete the comparisons between years or terroir and to be able to act according to the predictions of the model and the weather forecast. The major advantage of a service like this one is that it can allow one to monitor the vine water status in near-real-time (possibly every 5 days) on a large scale and at an affordable cost as the images are free.
The goal of this study was to verify the consistency of spatial and temporal monitoring of vine water status obtained with a model using S2 images. Images of vine fields of five wine estates have been processed with a previously developed model to predict their water status during the 2020 season. The feedback from winegrowers has been very positive according to their knowledge of their vine fields, the type of irrigation used or the weather data. Some limits have been highlighted like the necessity to well know the cultural practices and the vigour of the studied vine fields. The results obtained here are promising for future operational service in the Mediterranean region and further investigations will be done in other regions with different agronomic practices (e.g., grass, inter-row). The next step is to test this model systematically on many vine fields in real-time and asking winegrowers to go on the field and check the consistency of the water status estimation by measuring Ψstem or using a within-field notation with a mobile application like ApeX-Vigne (Pichon et al., 2021), for example.
The authors want to thank all the winegrowers implied in this study including the estates of Gourgazaud, La Baume, Les Homs, Chateau Canet, and Château de Cordes. They also thank Claire Quintard, Bénédicte Coulonges, Dorian Calvo, and Gabrielle Schmidt for the field measurements. The first author also thanks Audrey Choné, Florian Mouret, Mailys Lopes, and Mohanad Albughdadi for their help during the mapping process.
- Acevedo-Opazo, C., Tisseyre, B., Ojeda, H., Ortega-Farias, S., & Guillaume, S. (2008). Is it possible to assess the spatial variability of vine water status? OENO One, 42(4), 203. https://doi.org/10.20870/oeno-one.2008.42.4.811
- Addabbo, P., Focareta, M., Marcuccio, S., Votto, C., & Ullo, S. L. (2016). Contribution of Sentinel-2 data for applications in vegetation monitoring. ACTA IMEKO, 5(2), 44. https://doi.org/10.21014/acta_imeko.v5i2.352
- Balenzano, A., Mattia, F., Satalino, G., Lovergine, F. P., Palmisano, D., Peng, J., Marzahn, P., Wegmüller, U., Cartus, O., Dąbrowska-Zielińska, K., Musial, J. P., Davidson, M. W. J., Pauwels, V. R. N., Cosh, M. H., McNairn, H., Johnson, J. T., Walker, J. P., Yueh, S. H., Entekhabi, D., Jackson, T. J. (2021). Sentinel-1 soil moisture at 1 km resolution: a validation study. Remote Sensing of Environment, 263. https://doi.org/10.1016/j.rse.2021.112554
- Bellvert, J., Marsal, J., Mata, M., & Girona, J. (2012). Identifying irrigation zones across a 7.5-ha ‘Pinot noir’ vineyard based on the variability of vine water status and multispectral images. Irrigation Science, 30(6), 499–509. https://doi.org/10.1007/s00271-012-0380-y
- Bernardo, S., Dinis, L. T., Machado, N., & Moutinho-Pereira, J. (2018). Grapevine abiotic stress assessment and search for sustainable adaptation strategies in Mediterranean-like climates. A review. Agronomy for Sustainable Development, 38(6). https://doi.org/10.1007/s13593-018-0544-0
- Blanco-Cipollone, F., Lourenço, S., Silvestre, J., Conceição, N., Moñino, M. J., Vivas, A., Ferreira, M. I. (2017). Plant water status indicators for irrigation scheduling associated with iso-and anisohydric behavior: Vine and plum trees. Horticulturae, 3 (3). https://doi.org/10.3390/horticulturae3030047
- Borgogno-Mondino, E., Novello, V., Lessio, A., & de Palma, L. (2018). Describing the spatio-temporal variability of vines and soil by satellite-based spectral indices: A case study in Apulia (South Italy). International Journal of Applied Earth Observation and Geoinformation, 68(January), 42–50. https://doi.org/10.1016/j.jag.2018.01.013
- Bou Nader, K., Stoll, M., Rauhut, D., Patz, C.-D., Jung, R., Loehnertz, O., … Gomès, E. (2019). Impact of grapevine age on water status and productivity of Vitis vinifera L. cv. Riesling. European Journal of Agronomy, 104(January), 1–12. https://doi.org/10.1016/j.eja.2018.12.009
- Chaves, M. M., Santos, T. P., Souza, C. R., Ortuño, M. F., Rodrigues, M. L., Lopes, C. M., … Pereira, J. S. (2007). Deficit irrigation in grapevine improves water-use efficiency while controlling vigour and production quality. Annals of Applied Biology, 150(2), 237–252. https://doi.org/10.1111/j.1744-7348.2006.00123.x
- Choné, X., van Leeuwen, C., Dubourdieu, D., & Gaudillères, J. P. (2001). Stem Water Potential is a Sensitive Indicator of Grapevine Water Status. Annals of Botany, 87(4), 477–483. https://doi.org/10.1006/anbo.2000.1361
- CNES (2021). THEIA Platform. Retrieved May 27, 2021, from https://theia.cnes.fr
- Cogato, A., Pagay, V., Marinello, F., Meggio, F., Grace, P., & De Antoni Migliorati, M. (2019). Assessing the Feasibility of Using Sentinel-2 Imagery to Quantify the Impact of Heatwaves on Irrigated Vineyards. Remote Sensing, 11(23), 2869. https://doi.org/10.3390/rs11232869
- Cohen, Y., Gogumalla, P., Bahat, I., Netzer, Y., Ben-Gal, A., Lenski, I., … Helman, D. (2019). Can time series of multispectral satellite images be used to estimate stem water potential in vineyards? In Precision agriculture ’19 (pp. 445–451). The Netherlands: Wageningen Academic Publishers. https://doi.org/10.3920/978-90-8686-888-9_55
- Costa, J. M., Ortuño, M. F., & Chaves, M. M. (2007). Deficit Irrigation as a Strategy to Save Water: Physiology and Potential Application to Horticulture. Journal of Integrative Plant Biology, 49(10), 1421–1434. https://doi.org/10.1111/j.1672-9072.2007.00556.x
- Das, B., Sahoo, R. N., Pargal, S., Krishna, G., Verma, R., Viswanathan, C., Sehgal, V. K., Gupta, V. K. (2020). Evaluation of different water absorption bands, indices and multivariate models for water-deficit stress monitoring in rice using visible-near infrared spectroscopy. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 119104. https://doi.org/10.1016/j.saa.2020.119104
- Deloire, A., Pellegrino, A., & Rogiers, S. (2020). A few words on grapevine leaf water potential. IVES Technical Reviews, Vine and Wine, 2–3. https://doi.org/10.20870/IVES-TR.2020.3620
- Devaux, N., Crestey, T., Leroux, C., & Tisseyre, B. (2019). Potential of Sentinel-2 satellite images to monitor vine fields grown at a territorial scale. OENO One, 53(1), 51–58. https://doi.org/10.20870/oeno-one.2019.53.1.2293
- Di Gennaro, Salvatore F., Battiston, E., di Marco, S., Facini, O., Matese, A., Nocentini, M., … Mugnai, L. (2016). Unmanned Aerial Vehicle (UAV)-based remote sensing to monitor grapevine leaf stripe disease within a vineyard affected by esca complex. Phytopathologia Mediterranea. https://doi.org/10.14601/Phytopathol_Mediterr-18312
- Di Gennaro, Salvatore Filippo, Dainelli, R., Palliotti, A., Toscano, P., & Matese, A. (2019). Sentinel-2 validation for spatial variability assessment in overhead trellis system viticulture versus UAV and agronomic data. Remote Sensing, 11(21). https://doi.org/10.3390/rs11212573
- El Hajj, M.; Baghdadi, N.; Zribi, M.; Rodríguez-Fernández, N.; Wigneron, J.P.; Al-Yaari, A.; Al Bitar, A.; Albergel, C.; Calvet, J.-C. (2018) Evaluation of SMOS, SMAP, ASCAT and Sentinel-1 Soil Moisture Products at Sites in Southwestern France. Remote Sening., 10, 569. https://doi.org/10.3390/rs10040569
- Laroche-Pinel, E., Albughdadi, M., Duthoit, S., Chéret, V., Rousseau, J., & Clenet, H. (2021b). Understanding vine hyperspectral signature through different irrigation plans: A first step to monitor vineyard water status. Remote Sensing, 13(3), 1–31. https://doi.org/10.3390/rs13030536
- Laroche-Pinel, E., Duthoit, S., Albughdadi, M., Costard, A. D., Rousseau, J., Chéret, V., & Clenet, H. (2021a). Towards Vine Water Status Monitoring on a Large Scale Using Sentinel-2 Images. Remote Sensing, 13(9), 1837. https://doi.org/10.3390/rs13091837
- Ojeda, H., & Saurin, N. (2014). L’irrigation de précision de la vigne : méthodes, outils et stratégies pour maximiser la qualité et les rendements de la vendange en économisant de l’eau. Innovations Agronomiques, 38, 97–108.
- Pagay, V., Kidman, C. M. (2019). Evaluating Remotely-Sensed Grapevine (Vitis vinifera L.) Water Stress Responses across a Viticultural Region. Agronomy, 9 (11). https://doi.org/10.3390/agronomy9110682
- Parker, L. E., McElrone, A. J., Ostoja, S. M., & Forrestel, E. J. (2020). Extreme heat effects on perennial crops and strategies for sustaining future production. Plant Science, 295, 110397. https://doi.org/10.1016/j.plantsci.2019.110397
- Payan, J.-C. (2007). Controle du stress hydrique pour la gestion de l’irrigation en viticulture. In Euroviti.
- Pichon, L., Brunel, G., Payan, J. C., Taylor, J., Bellon-Maurel, V., & Tisseyre, B. (2021). ApeX-Vigne: experiences in monitoring vine water status from within-field to regional scales using crowdsourcing data from a free mobile phone application. Precision Agriculture, 22(2), 608–626. https://doi.org/10.1007/s11119-021-09797-9
- Poblete, T., Ortega-Farías, S., Moreno, M., & Bardeen, M. (2017). Artificial Neural Network to Predict Vine Water Status Spatial Variability Using Multispectral Information Obtained from an Unmanned Aerial Vehicle (UAV). Sensors, 17(2488), 1–17. https://doi.org/10.3390/s17112488
- Romero, M., Luo, Y., Su, B., & Fuentes, S. (2018). Vineyard water status estimation using multispectral imagery from an UAV platform and machine learning algorithms for irrigation scheduling management. Computers and Electronics in Agriculture, 147(February), 109–117. https://doi.org/10.1016/j.compag.2018.02.013
- Salva, G. (2016). Viticulture. L irrigation des vignes : notice à destination des utilisateurs et des autres. Revue Française d’oenologie, (274), 5. Retrieved from https://docplayer.fr/57595349-Viticulture-l-irrigation-des-vignes-notice-a-destination-des-utilisateurs-et-des-autres.html
- Santesteban, L. G., Miranda, C., Marín, D., Sesma, B., Intrigliolo, D. S., Mirás-Avalos, J. M., Escalona, J. M., Montoro, A., de Herralde, F., Baeza, P., Romero, P., Yuste, J., Uriarte, D., Martínez-Gascueña, J., Cancela, J. J., Pinillos, V., Loidi, M., Urrestarazu, J., Royo, J. B. (2019). Discrimination ability of leaf and stem water potential at different times of the day through a meta-analysis in grapevine (Vitis vinifera L.). Agricultural Water Management , 202–210. https://doi.org/10.1016/j.agwat.2019.04.020
- Scholander, P. F., Bradstreet, E. D., Hemmingsen, E. A., & Hammel, H. T. (1965). Sap Pressure in Vascular Plants: Negative hydrostatic pressure can be measured in plants. Science, 148(3668), 339–346. https://doi.org/10.1126/science.148.3668.339
- Sozzi, M., Kayad, A., Marinello, F., Taylor, J. A., & Tisseyre, B. (2020). Comparing vineyard imagery acquired from sentinel-2 and unmanned aerial vehicle (UAV) platform. Oeno One, 54(2), 189–197. https://doi.org/10.20870/oeno-one.2020.54.2.2557
- Suter, B., Triolo, R., Pernet, D., Dai, Z., van Leeuwen, C. (2019). Modeling Stem Water Potential by Separating the Effects of Soil Water Availability and Climatic Conditions on Water Status in Grapevine (Vitis vinifera L.). Frontiers in Plant Science, 10. https://doi.org/10.3389/fpls.2019.01485
- van Leeuwen, C., & Darriet, P. (2016). The Impact of Climate Change on Viticulture and Wine Quality. Journal of Wine Economics, 11(1), 150–167. https://doi.org/10.1017/jwe.2015.21
- van Leeuwen, C., Trégoat, O., Choné, X., Bois, B., Pernet, D., & Gaudillère, J.-P. (2009). Vine water status is a key factor in grape ripening and vintage quality for red Bordeaux wine. How can it be assessed for vineyard management purposes? OENO One, 43(3), 121. https://doi.org/10.20870/oeno-one.2009.43.3.798
- Weiss, M., Jacob, F., & Duveiller, G. (2020). Remote sensing for agricultural applications: A meta-review. Remote Sensing of Environment, 236(August 2019), 111402. https://doi.org/10.1016/j.rse.2019.111402
AttachmentsNo supporting information for this article