An artificial neural network approach to the estimation of stem water potential from frequency domain reflectometry soil moisture measurements and meteorological data

Abstract

ElsevierMartí Pérez, PC.; Gasque Albalate, M.; González Altozano, P. (2013). An artificial neural network approach to the estimation of stem water potential from frequency domain reflectometry soil moisture measurements and meteorological data. Computers and Electronics in Agriculture. 91:75-86. doi:10.1016Agriculture. 91:75-86. doi:10. /j.compag.2012.001. Dear Author, Please check your proof carefully and mark all corrections at the appropriate place in the proof (e.g., by using on-screen annotation in the PDF file) or compile them in a separate list. Note: if you opt to annotate the file with software other than Adobe Reader then please also highlight the appropriate place in the PDF file. To ensure fast publication of your paper please return your corrections within 48 hours.For correction or revision of any artwork, please consult http://www.elsevier.com/artworkinstructions.Any queries or remarks that have arisen during the processing of your manuscript are listed below and highlighted by flags in the proof. Click on the 'Q' link to go to the location in the proof. Location in articleQuery / Remark: click on the Q link to go Please insert your reply or correction at the corresponding line in the proof Q1Please confirm that given names and surnames have been identified correctly. Q2Kindly check whether the identification of corresponding author and the e-mail address are okay as typeset, and correct if necessary.Thank you for your assistance.Please check this box if you have no corrections to make to the PDF file Highlights " First step approach to estimate stem water potential from soil moisture and standard meteorological variables using ANNs. " Two principal components are enough to describe systematic variability of data. " Optimum input combination: temperature, relative humidity, solar radiation and soil moisture at 50 cm. " Artificial neural networks present higher accuracy than corresponding multi-linear regression models. COMPAG 2809No. of Pages 1, Model 5G , 1998;Hanson et al., 2000;Goldhamer and Fereres, 2001; 54 Dane and Topp, 2002;Intrigliolo and Castel, 2004;Jones, 2004). 55 In most of these, soil water status is considered a key factor for 56 planning irrigation doses (Campbell and Campbell, 1982).

Effects of long-term summer deficit irrigation on ‘Navelina’ citrus trees

“…It was also poor in total nitrogen (0.06%), available 121 potassium (0.42 meq K + 100g -1 ) and phosphorus (20.67 mg P kg -1 Olsen). A more detailed 122 description of the soil characteristics can be found in Martí et al (2013). The capacitance probes were properly installed within the active root system zone.…”

“…The first three sensors, located at 10, 30, and 50 cm of depth, 228 covered practically 90% of the active root system (estimated during probe installation), while 229 the fourth was outside of this zone (70, 80, and 60 cm depth in the probes for treatments T1, 230 T2, and the control, respectively). More details about the installation of these probes can be 231 found in Martí et al (2013). 232 233…”

“…Among others, the frequency domain reflectometry (FDR) probe, with multiple 71 depth capacitance sensors (Fares and Polyakov, 2006;Paltineanu and Starr, 1997), has shown 72 excellent performance so far. It is currently widely used for field applications as a decision-73 making tool for irrigation scheduling (Martí et al, 2013). 74…”

New Approaches for Teaching Irrigation Scheduling — Improving Understanding through Modeling Applications

“…Acevedo-Opazo et al [6] suggested a methodology to extrapolate pre-dawn leaf water potential using linear combinations of ancillary data. Recently, Martí et al [7] used artificial neural networks to estimate stem water potential from soil moisture measurements and standard climatic records.…”

The identification of relationships between selected honey parameters by extracting the contribution of independent variables in a neural network model

“…For solving regression tasks, the multilayer perceptron (MLP) is mainly used. Shafiee et al [14] have used MLP for prediction of some chemical honey parameters from color features, Marti et al [15] for the estimation of stem water potential, Maulidiani et al [16] for modeling the relationship between the bioactive compounds in Pegaga extract and antioxidant activity and Xi et al [17] for modeling effects of pressure, liquid/solid ratio and ethanol concentration on the total phenolic content of green tea extracts. For classification tasks, the self-organizing maps (SOM) or MLP networks are commonly used.…”