scholarly journals Grapevine Phenology in Four Portuguese Wine Regions: Modeling and Predictions

2020 ◽  
Vol 10 (11) ◽  
pp. 3708
Author(s):  
Samuel Reis ◽  
Helder Fraga ◽  
Cristina Carlos ◽  
José Silvestre ◽  
José Eiras-Dias ◽  
...  
Keyword(s):  
High Performance ◽  
Mean Squared Error ◽  
Cultural Practices ◽  
Meteorological Data ◽  
Efficiency Coefficient ◽  
Phenological Stages ◽  
Predictive Capacity ◽  
Phenological Model ◽  
Parameter Estimations ◽  
Wine Region

Phenological models applied to grapevines are valuable tools to assist in the decision of cultural practices related to winegrowers and winemakers. The two-parameter sigmoid phenological model was used to estimate the three main phenological stages of the grapevine development, i.e., budburst, flowering, and veraison. This model was calibrated and validated with phenology data for 51 grapevine varieties distributed in four wine regions in Portugal (Lisboa, Douro, Dão, and Vinhos Verdes). Meteorological data for the selected sites were also used. Hence, 153 model calibrations (51 varieties × 3 phenological stages) and corresponding parameter estimations were carried out based on an unprecedented comprehensive and systematized dataset of phenology in Portugal. For each phenological stage, the centroid of the estimated parameters was subsequently used, and three generalized sigmoid models (GSM) were constructed (budburst: d = −0.6, e = 8.6; flowering: d = −0.6, e = 13.7; veraison: d = −0.5, e = 13.2). Centroid parameters show high performance for approximately 90% of the varieties and can thereby be used instead of variety-specific parameters. Overall, the RMSE (root-mean-squared-error) is < 7 days, while the EF (efficiency coefficient) is > 0.5. Additionally, according to other studies, the predictive capacity of the models for budburst remains lower than for flowering or veraison. Furthermore, the F-forcing parameter (thermal accumulation) was evaluated for the Lisboa wine region, where the sample size is larger, and for the varieties with model efficiency equal to or greater than 0.5. A ranking and categorization of the varieties in early, intermediate, and late varieties was subsequently undertaken on the basis of F values. These results can be used to more accurately monitor and predict grapevine phenology during a given season, thus supporting decision making in the Portuguese wine sector.

scholarly journals Grapevine Phenology in Four Portuguese Wine Regions: Modeling and Predictions

2021 ◽  
Author(s):  
Samuel Reis ◽  
Helder Fraga ◽  
Cristina Carlos ◽  
José Silvestre ◽  
José Eiras-Dias ◽  
...  
Keyword(s):  
High Performance ◽  
Mean Squared Error ◽  
Cultural Practices ◽  
Meteorological Data ◽  
Efficiency Coefficient ◽  
Short Term ◽  
Phenological Stages ◽  
Predictive Capacity ◽  
Phenological Model ◽  
Wine Region

&lt;p&gt;Phenological models applied to grapevines are valuable tools to assist in the decision of cultural practices related to winegrowers and winemakers. The two-parameter sigmoid phenological model was used to estimate the three main phenological stages of the grapevine development, i.e., budburst, flowering, and veraison. This model was calibrated and validated with phenology data for 51 grapevine varieties distributed in four wine regions in Portugal (Lisboa, Douro, D&amp;#227;o, and Vinhos Verdes). Meteorological data for the selected sites were also used. Hence, 153 model calibrations (51 varieties &amp;#215; 3 phenological stages) and corresponding parameter estimations were carried out based on an unprecedented comprehensive and systematized dataset of phenology in Portugal. For each phenological stage, the centroid of the estimated parameters was subsequently used, and three generalized sigmoid models were constructed (budburst: d =&amp;#8722;0.6, e = 8.6; flowering: d = &amp;#8722;0.6, e = 13.7; veraison: d = &amp;#8722;0.5, e = 13.2). Centroid parameters show high performance for approximately 90% of the varieties and can thereby be used instead of variety-specific parameters. Overall, the RMSE (root-mean-squared-error) is &lt; 7 days, while the EF (efficiency coefficient) is &gt; 0.5. Additionally, according to other studies, the predictive capacity of the models for budburst remains lower than for flowering or veraison. Furthermore, the F-forcing parameter (thermal accumulation) was evaluated for the Lisboa wine region, where the sample size is larger, and for the varieties with model efficiency equal to or greater than 0.5. A ranking and categorization of the varieties in early, intermediate, and late varieties was subsequently undertaken on the basis of F values. In this way, these results of the present study will be incorporated on a web platform, where the sigmoid model must convey valuable information regarding the development/evolution of the vineyard with short-term predictions.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Keywords: &lt;/strong&gt;grapevine; phenology modeling; sigmoid model; wine regions; short-term predictions; Portugal&lt;/p&gt;

scholarly journals Relationship Between Phenological and Meteorological Data as an Important Input Into Spring Barley Phenological Model

2019 ◽  
Vol 67 (3) ◽  
pp. 679-688
Author(s):  
Lenka Hájková ◽  
Martin Možný ◽  
Věra Kožnarová ◽  
Lenka Bartošová ◽  
Zdeněk Žalud
Keyword(s):  
Meteorological Data ◽  
Spring Barley ◽  
Meteorological Parameter ◽  
Grain Filling ◽  
Phenological Stage ◽  
Phenological Stages ◽  
Phenological Model ◽  
Average Value ◽  
Precipitation Total ◽  
Grain Filling Period

In this study, phenological and meteorological data have been used to interpret the relationship and influence of weather on current phenological stages of spring barley. The analyses were focused mainly on the stages closely connected with yield and grain filling period – tillering (BBCH 21), heading (BBCH 55) and yellow ripeness (BBCH 85). The aims of this paper were to: (1) calculate the trend in phenological development of spring barley from CHMI phenological stations in period 1991 – 2012 at different climatic zones; (2) evaluate the trend in number of days between phenological stages; (3) evaluate the sums of growing degree days above threshold above 5 °C (GDD) and precipitation totals to phenophase onset calculated since the phenological stage of emergence (BBCH 10); (4) calculate Pearson’s correlation coefficient (PCC) between phenological stage and meteorological parameter. The highest positive PCC was found between GDD and phenological stages of heading and yellow ripeness at Doksany and Strážnice stations situated in lowlands. The average value of GDD to phenological stage heading is within the range from 418.4 to 500.1 °C. The sums of precipitation totals fluctuate from 73.9 mm (Doksany station) to 123.2 mm (Chrastava station). The results of this study suggest that GDD can be a more suitable parameter for phenological model of spring barley development than precipitation total.

scholarly journals Modelling the phenological development of cv. Touriga Nacional and Encruzado in the Dão Wine Region, Portugal

OENO One ◽  
2021 ◽  
Vol 55 (3) ◽  
pp. 337-352
Author(s):  
Pedro Rodrigues ◽  
Vanda Pedroso ◽  
Carla Henriques ◽  
Ana Matos ◽  
Samuel Reis ◽  
...  
Keyword(s):  
Decision Support ◽  
High Performance ◽  
Model Development ◽  
Decision Support Tool ◽  
Weather Conditions ◽  
Sigmoid Function ◽  
Phenological Stages ◽  
Support Tool ◽  
Berry Quality ◽  
Wine Region

The grapevine vegetative cycle, which is morphologically described by its phenological stages, is strongly determined by weather conditions. Phenological models are widely applied in viticulture and are based on the assumption that air temperature is the preponderant environmental factor which determines vine development. In this study, phenological development models (PDMs) were calibrated and validated to simulate several intermediate stages between budbreak and veraison for cv. Touriga Nacional (TN) and cv. Encruzado (EN) winegrape varieties, which are widely grown in the Dao Wine Region, Portugal. These are thermal models, with which the daily sum of the rate of forcing (R) was calculated using a sigmoid function. For this purpose, a high-quality and comprehensive dataset was used which combines phenology data and weather station data in several vineyard sites spread over the region. The model showed an overall high performance (global RMSE of 5.4 days for EN and 5.0 days for TN), although it depended on the phenological stage and variety. The RMSE ranged from 3.2 to 6.2 for TN, and from 3.9 to 6.8 for EN. For both varieties and in all phenological stages, the RMSE was significantly lower than the standard deviation of the phenological observations. For TN, the model efficiency was greater than 0.71 for all phenological stages. In future studies, these models will be combined with specific models that simulate the evolution of winegrape berry quality indicators commonly used for harvest decision support. The relatively low complexity of the selected PDMs enables their use as a crop management and decision support tool. To our knowledge, no previous studies have been carried out on either of these two varieties and their intermediate phenological timings. The present study is an illustration of conceivable model development under diverse environmental conditions, thus allowing similar approaches to be adopted in other wine regions on a worldwide scale.

Prognosis of water levels in a moor groundwater system influenced by hydrology and water extraction using an artificial neural network 

2021 ◽  
Author(s):  
Sascha Flaig ◽  
Timothy Praditia ◽  
Alexander Kissinger ◽  
Ulrich Lang ◽  
Sergey Oladyshkin ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Mean Squared Error ◽  
Meteorological Data ◽  
Water Levels ◽  
Typical Structure ◽  
Water Abstraction ◽  
Physical Knowledge ◽  
Artificial Neural ◽  
The Impact

&lt;p&gt;In order to prevent possible negative impacts of water abstraction in an ecologically sensitive moor south of Munich (Germany), a &amp;#8220;predictive control&amp;#8221; scheme is in place. We design an artificial neural network (ANN) to provide predictions of moor water levels and to separate hydrological from anthropogenic effects. As the moor is a dynamic system, we adopt the &amp;#8222;Long short-term memory&amp;#8220; architecture.&lt;/p&gt;&lt;p&gt;To find the best LSTM setup, we train, test and compare LSTMs with two different structures: (1) the non-recurrent one-to-one structure, where the series of inputs are accumulated and fed into the LSTM; and (2) the recurrent many-to-many structure, where inputs gradually enter the LSTM (including LSTM forecasts from previous forecast time steps). The outputs of our LSTMs then feed into a readout layer that converts the hidden states into water level predictions. We hypothesize that the recurrent structure is the better structure because it better resembles the typical structure of differential equations for dynamic systems, as they would usually be used for hydro(geo)logical systems. We evaluate the comparison with the mean squared error as test metric, and conclude that the recurrent many-to-many LSTM performs better for the analyzed complex situations. It also produces plausible predictions with reasonable accuracy for seven days prediction horizon.&lt;/p&gt;&lt;p&gt;Furthermore, we analyze the impact of preprocessing meteorological data to evapotranspiration data using typical ETA models. Inserting knowledge into the LSTM in the form of ETA models (rather than implicitly having the LSTM learn the ETA relations) leads to superior prediction results. This finding aligns well with current ideas on physically-inspired machine learning.&lt;/p&gt;&lt;p&gt;As an additional validation step, we investigate whether our ANN is able to correctly identify both anthropogenic and natural influences and their interaction. To this end, we investigate two comparable pumping events under different meteorological conditions. Results indicate that all individual and combined influences of input parameters on water levels can be represented well. The neural networks recognize correctly that the predominant precipitation and lower evapotranspiration during one pumping event leads to a lower decrease of the hydrograph.&lt;/p&gt;&lt;p&gt;To further demonstrate the capability of the trained neural network, scenarios of pumping events are created and simulated.&lt;/p&gt;&lt;p&gt;In conclusion, we show that more robust and accurate predictions of moor water levels can be obtained if available physical knowledge of the modeled system is used to design and train the neural network. The artificial neural network can be a useful instrument to assess the impact of water abstraction by quantifying the anthropogenic influence.&lt;/p&gt;

scholarly journals Serum N-Glycomic Biomarkers Predict Treatment Escalation in Inflammatory Bowel Disease

2021 ◽  
Author(s):  
Archana Shubhakar ◽  
Bas C Jansen ◽  
Alex T. Adams ◽  
Karli R. Reiding ◽  
Nicholas T. Ventham ◽  
...  
Keyword(s):  
Inflammatory Bowel Disease ◽  
Bowel Disease ◽  
High Performance ◽  
Clinical Decision Making ◽  
Clinical Decision ◽  
Predictive Capacity ◽  
Cox Proportional Hazard Models ◽  
Independent Replication ◽  
Inflammatory Bowel ◽  
Treatment Escalation

Abstract A blood-based prognostic biomarker to guide clinical decision-making at diagnosis of inflammatory bowel disease (IBD) would be immensely helpful. We investigated a composite serum N-glycomic biomarker to predict future disease course in 244 newly diagnosed IBD patients. Forty-seven individual glycan peaks were analysed using ultra-high performance liquid chromatography identifying 105 glycoforms from which 24 derived glycan traits were calculated. Multivariable logistic regression was performed to determine associations of derived glycan traits with disease. Cox proportional hazard models were used to predict treatment escalation from first-line treatment to biologics or surgery (hazard ratio (HR) 25.9, p = 1.1×10− 12; 95% confidence interval (CI), 8.52–78.78). Application to an independent replication cohort of 54 IBD patients yielded a HR of 5.1 (p = 1.1×10− 5; 95% CI, 2.54–10.1). These data demonstrate the predictive capacity of serum N-glycan biomarkers and represent a step towards personalized medicine in IBD.

scholarly journals Effect of unfavourable population structure on estimates of heritability, systematic effects and breeding values

2008 ◽  
Vol 51 (6) ◽  
pp. 601-610
Author(s):  
A. P. Kominakis
Keyword(s):  
Population Structure ◽  
Mean Squared Error ◽  
Heritability Estimate ◽  
Breeding Values ◽  
Parameter Estimations ◽  
Squared Error ◽  
Heritability Estimation ◽  
Population Structures ◽  
Systematic Effects ◽  
The Impact

Abstract. Empirical estimations of heritability, systematic effects and predictions of sires’ breeding values (BVs) were obtained under various population structures for simulated populations consisted of n = 400 animals in 5 herds for a trait of medium heritability (h2 = 0.30). An infinitesimal additive genetic animal model was assumed while simulating data. Population structure was varied to allow for good and poor connectedness across herds and (non)random association between the genetic and the environmental effects. The impact of the various population structures on the parameter estimation(s) was assessed using Mean Squared Error (MSE) and Pearson’s correlations. Allowing sires to have progenies in more than one herd (good herd connectedness) and random use of sires across herds generally resulted in good parameter estimations. Poor connectedness significantly affected herd effects estimation and BV prediction but not heritability estimation as long as random usage of sires across environments was guaranteed. Selective use of the best sires in the best herds along with poor connectedness resulted in poorest estimations of all parameters examined. In the latter case, heritability was seriously underestimated (h2 = 0.06) while highest error, lowest accuracies for the BVs and a remarkable underestimation of the genetic gain were observed. Use of reference sires on a natural mating basis to create genetic links between herds has served a good solution for both heritability and BVs estimation under unfavorable structure. Mating 0.25 of the herd ewes with reference sires resulted in a heritability estimate close to the simulated one. Significantly better estimates of systematic effects and BVs were, however, obtained when 0.5 of the herd ewes were mated by reference sires.

Two-Run Genetic Programming for Predicting Slump Flow of Concrete

2014 ◽  
Vol 590 ◽  
pp. 321-325
Author(s):  
Li Chen ◽  
Chang Huan Kou ◽  
Kuan Ting Chen ◽  
Shih Wei Ma
Keyword(s):  
Nonlinear Systems ◽  
Genetic Programming ◽  
High Performance ◽  
Mean Squared Error ◽  
High Performance Concrete ◽  
Local Optimum ◽  
Root Mean Squared Error ◽  
Squared Error ◽  
Slump Flow ◽  
Low Dimensional

A two-run genetic programming (GP) is proposed to estimate the slump flow of high-performance concrete (HPC) using several significant concrete ingredients in this study. GP optimizes functions and their associated coefficients simultaneously and is suitable to automatically discover relationships between nonlinear systems. Basic-GP usually suffers from premature convergence, which cannot acquire satisfying solutions and show satisfied performance only on low dimensional problems. Therefore it was improved by an automatically incremental procedure to improve the search ability and avoid local optimum. The results demonstrated that two-run GP generates an accurate formula through and has 7.5 % improvement on root mean squared error (RMSE) for predicting the slump flow of HPC than Basic-GP.

scholarly journals Influence of pedoclimatic factors on grapevine performance and berry ripening: Preliminary results of Chardonnay grown in Friuli Venezia Giulia region, Italy

2019 ◽  
Vol 15 ◽  
pp. 01010
Author(s):  
L. Bavaresco ◽  
J. Lemaignen ◽  
E. Neethling ◽  
C. Squeri ◽  
C. Coulon-Leroy
Keyword(s):  
Cultural Practices ◽  
Wine Industry ◽  
Stem Water Potential ◽  
Study Objective ◽  
Berry Ripening ◽  
Accurate Perception ◽  
Stem Water ◽  
Wine Region ◽  
Terrain Features ◽  
Friuli Venezia Giulia

The concept of terroir is widely used in the wine industry, and many studies are undertaken to better understand the influence of local terrain features on produced wines. In this context, this study monitored nine Chardonnay plots in the Fruili wine region of Italy to gather information on grapevine growth and berry ripening. The study objective was to define viticulture potentialities allowing to identify future strategies for the production of a “single vineyard” Chardonnay wine at the Vie di Romans estate, Italy. During the growing season of 2017, relations were studied between terrain features and field observations on vine phenology and grape ripening. Stem water potential and gas exchange measurements were also effectuated. Results show that there are significant differences between the studied plots. Earliness of the grapevine cycle between the plots has been less evident to determine than the variation in berry characteristics. The latter were linked with identified pedoclimatic units, but the effect of the cultural practices should not be overlooked. The study should be perceived as a first monitoring campaign, highlighting the important differences between study plots. Further investigation in the following seasons should give a more accurate perception of individual plot characteristics and their impacts.

Extended Modeling of Vertical Axis Motion Dynamics of VTOL Vehicle

2015 ◽  
Vol 789-790 ◽  
pp. 883-888 ◽  
Author(s):  
Wojciech Janusz ◽  
Roman Czyba ◽  
Grzegorz Szafrański ◽  
Michał Niezabitowski
Keyword(s):  
Vehicle Dynamics ◽  
High Performance ◽  
Vertical Axis ◽  
Open Loop ◽  
Indirect Measurements ◽  
Parameter Estimations ◽  
Landing Platform ◽  
Standard Configuration ◽  
Aerial Vehicle ◽  
Practical Model

Development of a reliable high-performance multirotor unmanned aerial vehicle (UAV) requires an accurate and practical model of the vehicle dynamics. This paper describes the process and results of the dynamic modeling of an unmanned aerial platform known as quadrotor. To model a vehicle dynamics, elementary physical and aerodynamical principles has been employed. Parameter estimations, from a UAV design have been obtained through direct and indirect measurements. In addition to standard configuration of VTOL (Vertical Take-Off and Landing) platform, the amortized landing gear, modeled as spring-damper system, has been added. The resulting model has been implemented in a simulation environment under MATLABs toolbox, SIMULINK. Some numerical results are presented to illustrate response of the open loop system to specific commands.

scholarly journals Missing Value Imputation using XGboost for Label-Free Mass Spectrometry-Based Proteomics Data

2021 ◽  
Author(s):  
Jian Song ◽  
Changbin Yu
Keyword(s):  
Mass Spectrometry ◽  
High Performance ◽  
Missing Values ◽  
Mean Squared Error ◽  
Learning Algorithm ◽  
Pearson Correlation ◽  
Data Matrix ◽  
Label Free ◽  
Proteomics Data ◽  
Benchmark Datasets

AbstractThe label-free mass spectrometry-based proteomics data inevitably suffer from the problem of missing values. The existence of missing values prevents the downstream analyses which need a complete data matrix. Our motivation is to introduce the state-of-art machine learning algorithm XGboost to realize a method of imputation which can improve the accuracy of imputation. But in practical, XGboost has many parameters need to be tuned to deliver on its potential high performance. Although cross validation may find the best parameters, it is much time-consuming. Alternatively, we empirically determined the parameters to two kinds of base learners of XGboost. To explore the robustness and performance of XGboost based imputation with predetermined parameters, we conducted tests on three benchmark datasets. As a comparative, six common imputation methods were also experimented in terms of normalized root mean squared error and Pearson correlation coefficient. The comparative experimental results indicated that the XGboost based imputation method using the linear base learner is competitive to or out-performs its competitors, including the random forest based imputation, by achieving smaller imputation errors and better structure preservation under the empirical parameters for the three benchmark datasets.

Sign in / Sign up

Export Citation Format

Share Document