Limitations and usefulness of maximum daily shrinkage (MDS) and trunk growth rate (TGR) indicators in the irrigation scheduling of table olive trees

Elsevier Editorial System(tm) for Agricultural Water Management Manuscript Draft Manuscript Number: Title: Limitations and usefulness of Maximum Daily Shrinkage (MDS) and trunk growth rate (TGR) indicators in the irrigation scheduling of table olive trees. Article Type: SI: Plants Water Saving Keywords: Olea europaea; trunk diameter fluctuations; water potential; water relations. Corresponding Author: Dr. Alfonso Moriana, Ph D Corresponding Author's Institution: Sevilla University First Author: Ignacio Girón Order of Authors: Ignacio Girón; Mireia Corell; Alejandro Galindo; Arturo Torrecillas; Félix Moreno; Alfonso Moriana, Ph D Abstract: Maximum daily trunk shrinkage (MDS) is the most popular indicator derived from trunk diameter fluctuations in most fruit trees and has been reported to be one of the earliest signs in the detection of water stress. However, in some species such as olive trees (Olea europaea L), MDS does not usually change in water stress conditions and trunk growth rate (TGR) has been suggested as better indicator. Most of this lack of sensitivity to drought conditions has been related to the relationship between the MDS and the water potential. This curvilinear relationship produces an uncertain zone were great variations of water potential do not imply any changes of MDS. The MDS signal, the ratio between measured MDS and estimated MDS with full irrigation, has been thought to be a better indicator than MDS, as it reduces the effect of the environment. New methodologies for estimation of the MDS signal in olive trees have been recently suggested. On the other hand, though literature results suggest an effect of environment in TGR values, there are not clear relationship between this indicator and meteorological data. The aims of this work are, on one hand, to study the improvements of the baseline approach in the MDS signal and on the other study the influence of several meteorological variables in TGR. Three years' data from an irrigation experiment were used in to carry out the MDS analysis and six years' data for full irrigated trees were used for TGR study. The comparison between MDS vs water potential and MDS signal vs water potential presented a great scattering in both relationships. However, in the interval of water potential between -1.4 and -2MPa, the MDS signal presented a clear increase, which was not identified in the relationship of MDS vs. water potential. It is likely that the seasonal estimation of the baseline would provide a better adjustment of the MDS signal in relation to the water potential and could be useful at the beginning of the water stress period. On the other hand, TGR was affected significantly for the increment of the daily average vapour pressure deficit (VPD) of the previous day and this relationship was affected for the fruit load level. Suggested Reviewers: Facundo Vita [email protected] Mario Gracia-Petillo [email protected] Ricardo Gucci [email protected] Opposed Reviewers: B. Clothier Editor Agricultural Water Management Dear Dr.Clothier: We should be grateful if you would consider the attached manuscript entitled “LIMITATIONS AND USEFULNESS OF MAXIMUM DAILY SHRINKAGE (MDS) AND TRUNK GROWTH RATE (TGR) INDICATORS IN THE IRRIGATION SCHEDULING OF TABLE OLIVE TREES.” for publication in the special issue of Plant Water Saving of the Journal Agricultural Water Management. Our work reports data that provide information about the management of trunk diameter fluctuations indicators in table olive trees. In one hand, we have reported the relationship between maximum daily shrinkage (MDS) signal and water potential using the approach of Corell et al (2013) in order to verify if it is useful in moderate water stress conditions. On the other hand, we have compared meteorological data of 5 different high fruit load seasons with trunk diameter fluctuations (TGR) in order to study the influence of environment in this indicator. The present work suggests that MDS signal cold be partially useful in moderate water stress conditions and TGR is affected for the increment in the vapour pressure deficit (VPD). All the authors have read the manuscript and approved it for publication. Sincerely yours Alfonso Moriana Cover Letter Headlights MDS signal greater than 1.1 identified moderate water stress conditions MDS signal increase from 1.1 is not linearity related with higher water stress levels MDS signal increased in low fruit load with no relation with water stress TGR was related with the increment of daily average VPD of the previous day TGR vs VPD relationship was affected for the fruit load in full irrigated orchards *Highlights (for review) 1 Limitations and usefulness of Maximum Daily Shrinkage (MDS) and trunk growth 1 rate (TGR) indicators in the irrigation scheduling of table olive trees. 2 3 I.F. Giróna,b M. Corellc, b, A. Galindod, A. Torrecillasd, F. Morenoa,b, A. Morianac,b* 4 5 a Instituto de Recursos Naturales y Agrobiología (CSIC), P.O. Box 1052, E-41080 6 Sevilla, Spain. 7 b Unidad Asociada al CSIC de Uso sostenible del suelo y el agua en la agricultura (US-8 IRNAS). Crta de Utrera Km 1, 41013, Sevilla, Spain. 9 c Departamento de Ciencias Agroforestales, ETSIA, Universidad de Sevilla, Crta de 10 Utrera Km 1. 41013 Sevilla, Spain. 11 d Dpto. Riego. Centro de Edafología y Biología Aplicada del Segura (CSIC). P.O. Box 12 164. E-301000 Espinardo, Murcia, Spain 13 *Corresponding author: [email protected] Phone: (+34)954486456; Fax: 14 (+34)954486436 15 16 17 18 19 20 21 22 23 24 *Manuscript Click here to download Manuscript: giron et al.docx Click here to view linked References 2 Abstract 25 Maximum daily trunk shrinkage (MDS) is the most popular indicator derived from 26 trunk diameter fluctuations in most fruit trees and has been reported to be one of the 27 earliest signs in the detection of water stress. However, in some species such as olive 28 trees (Olea europaea L), MDS does not usually change in water stress conditions and 29 trunk growth rate (TGR) has been suggested as better indicator. Most of this lack of 30 sensitivity to drought conditions has been related to the relationship between the MDS 31 and the water potential. This curvilinear relationship produces an uncertain zone were 32 great variations of water potential do not imply any changes of MDS. The MDS signal, 33 the ratio between measured MDS and estimated MDS with full irrigation, has been 34 thought to be a better indicator than MDS, as it reduces the effect of the environment. 35 New methodologies for estimation of the MDS signal in olive trees have been recently 36 suggested. On the other hand, though literature results suggest an effect of environment 37 in TGR values, there are not clear relationship between this indicator and 38 meteorological data. The aims of this work are, on one hand, to study the improvements 39 of the baseline approach in the MDS signal and on the other study the influence of 40 several meteorological variables in TGR. Three years’ data from an irrigation 41 experiment were used in to carry out the MDS analysis and six years’ data for full 42 irrigated trees were used for TGR study. The comparison between MDS vs water 43 potential and MDS signal vs water potential presented a great scattering in both 44 relationships. However, in the interval of water potential between -1.4 and -2MPa, the 45 MDS signal presented a clear increase, which was not identified in the relationship of 46 MDS vs. water potential. It is likely that the seasonal estimation of the baseline would 47 provide a better adjustment of the MDS signal in relation to the water potential and 48 could be useful at the beginning of the water stress period. On the other hand, TGR was 49 3 affected significantly for the increment of the daily average vapour pressure deficit 50 (VPD) of the previous day and this relationship was affected for the fruit load level. 51 52 Keywords: Olea europaea, trunk diameter fluctuations, water potential, water 53 relations. 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 4 Introduction 75 Trunk diameter fluctuations (TDF) are a daily cycle of shrinking and swelling of the 76 trunk that have been reported since the 60’s (Ortuño et al., 2010). The development of 77 sensors and dataloggers during the 90’s allowed a re-discovery of the usefulness of 78 these indicators in irrigation scheduling. Research works about drought response of 79 TDF indicators in fruit trees and even automatic irrigation scheduling based on these 80 indicators have been reported (i.e. “Pepista”, Huguet et al., 1992). 81 Several types of indicators can be obtained from the daily TDF curves. The most 82 common and early sign of water stress in most fruit trees is the maximum daily 83 shrinkage (MDS) (Ortuño et al. 2010). The increase of MDS compared to fully irrigated 84 trees was reported, from the first works, as an indicator of water stress conditions 85 (Klepper et al, 1971). However, the increase in MDS is also related to the evaporative 86 demand (Herzog et al., 1995). Thus, evaporative demand is an interference of this 87 indicator that reduces its usefulness in commercial orchards. In order to reduce this 88 limitation, Goldhamer and Fereres (2001) suggested the MDS signal: the ratio between 89 measured MDS and MDS with full irrigation. Fully irrigated conditions could be 90 estimated from baseline equations, where MDS is related to a meteorological variable, 91 such as reference evapotranspiration (ETo), vapour pressure deficit (VPD) or 92 temperature (Ortuño et al, 2010; Fernández and Cuevas. 2010). 93 The usefulness of MDS, however, is not the same in all fruit species. In young 94 olive trees, MDS was not affected by water stress, even when gas exchange was reduced 95 (Moriana and Fereres, 2002). This lack of response has been reported in mature olive 96 trees and in different cultivars and conditions (Moriana et al., 2003; Moriana et al., 97 2010; Fernández et al., 2011). Moriana et al (2010) suggested that the absence of 98 response to water stress in MDS is related to the pattern of this indicator during a 99 5 drought cycle. The relationship between MDS and water potential is curvilinear in all 100 fruit trees, showing an initial increase of MDS with the reduction of the water potential 101 until reaching a maximum value, and then MDS values decrease as the severity of water 102 stress continues to increase (Ortuño et al, 2010). This relationship presented the highest 103 MDS values in olive trees (maximum around 0.8-1mm) (Moriana et al., 2000) and the 104 first linear phase, until the maximum MSD values, has been considered to be caused by 105 variations in the conditions of the evaporative demand (Pérez-López et al., 2013). Since 106 the MDS values during summer in fully irrigated olive trees were around the maximum, 107 moderate water stress conditions would be in the uncertain zone were clear differences 108 of water potential (between -1.4MPa and -2MPa) presented similar MDS values. In 109 addition, Moriana et al (2013) reported greater values of MDS in fully irrigated 110 conditions for trees which were deficit irrigated in the previous season than in trees with 111 full irrigation. The MDS baseline is likely to reduce the influence of the environment on 112 this indicator but it is not known if it would be a reliable indicator in moderate water 113 stress conditions. Corell et al (2013) recently reported on a methodology for the 114 estimated MDS baseline at the beginning of the season which could reduce some of the 115 limitations presented above. 116 These limitations in the usefulness of MDS in olive trees have produced that 117 other indicators such as trunk growth rate (TGR) have been considered for irrigation 118 scheduling (Moriana et al., 2013). TGR is clearly affected for the fruit load and during 119 pit hardening period in mature trees almost no growth is detected (Moriana et al., 2003). 120 However, even in these conditions, TGR in full irrigated olive trees is very variable and 121 extremely negative values are measured (Moriana et al., 2013). Such response suggests 122 an environmental effect which has been poorly described in olive trees. 123 124 12 1.6MPa. This pattern changed at around -1.8MPa when the maximum MDS was 274 reached. Then, there was also a clear trend for MDS to decrease with values lower 275 than -2MPa. The lowest  Control values were in the range of 600-800m, similar 276 values to those from RDI treatments and close to the maximum MDS measured. 277 The data of Figure 2 were grouped in the same  intervals as in the previous 278 Figure. An MDS signal equal to 1 represents a theoretical value of conditions with full 279 irrigation. Figure 4 shows a confidence interval of around 10%, therefore MDS signal 280 values from 0.9 to 1.1 could be included in the group with full irrigation. All the Control 281 data are within the interval 1-1.1 of the MDS signal, though the water potential changed 282 from near -1.4MPa to slightly under -1.8MPa. There is also a clear differentiation 283 between data for 2011 and the rest of seasons. Most of the MDS signal data in this 284 season are above 1.1, even in the Control group, though  values were near -1.4MPa 285 (Figure 4). On the other hand, all of the RDI data from -1.6 to -2MPa clearly presented 286 an MDS signal higher than 1.1, with maximum average values around 1.4. When the 287 values were lower than -2MPa, MDS signals were under 1.1 (Figure 4). 288 Relationship between trunk growth rate (TGR) and environment 289 The best relationship between TGR and meteorological data for each season is 290 presented at Table 3. Most of the relationships calculated were not significant (data not 291 shown). The ones presented here are only the best four for each season and orchard in 292 order to improve the clarity of results. None of the regressions that included absolute 293 value of the meteorological data and the daily value of TGR were significant (data not 294 shown). When TGR values were related with the increment of each meteorological 295 variable in some years the regression was significant, but they were still very poor (data 296 not shown). Only when these increments were related with the TCT of the next day the 297 signification was improved. The data of the best relationships between previous 298 13 meteorological data and TGR for each year are presented at Table 3, only data of the 299 year 2008 is not presented. Average daily relative humidity ((-1)RHav), temperature 300 (average or maximum) and daily average vapour pressure deficit ((-1)VPDav) were 301 the best relationship with TGR. Only (-1)VPDav and (-1)RHav were one of the best 302 in all the years considered (Table 3). Determination coefficient in these variables 303 changed from 0.34 to 0.61 in (-1)VPDav and from 0.2 to 0,52 in (-1)RHav (Table 3). 304 None of the other variables or any multivariable equations improved the results of these 305 two. In all the relationships of Table 3, TGR decreased with an increase of the 306 evaporative demand. The slope in the (-1)VPDav vs TGR relationship was the greatest 307 in all the seasons considered (between -48.6 to -65.5 mdía-1KPa-1, while in the others 308 from 3.0 in (-1) RHav to -18.8 (-1)Tav (increment of the day before in average 309 temperature). 310 Similar relationships were obtained when data from a different near orchard was 311 considered (orchard 7*7; 2012 season, Table 3). Accuracy of the equation was 312 improved in this orchard, but (-1)VPDav was again the best variable. This equation 313 explained the 75% of the data variability and the slope was 4 times greater than the rest 314 of the equations (Table 3). The equation of this orchard was significantly different from 315 the ones of the 7*5 orchard. 316 Although the (-1)VPDav vs TGR relationships presented different slopes 317 between years (Table 3), such differences were not significant (Fig. 5) for the 7*5 318 orchard. The equation that considered the pool data presented a R2 around 0.45. The 319 slope of this equation suggests an important effect of the VPD in the TGR (around 55 320 m día-1 per KPa). This equation was significantly different from the ones obtained in 321 7*7 orchard. However, in the interval ±1 KPa of VPD, where most of the data are 322 presented, both equations are very similar (Fig. 5). 323 14 The accuracy of the equations in the 7*5 orchard was very different between 324 seasons (Table 3). When R2 is related with the fruit yield of each season, a clear trend to 325 lower influence of VPD with an increase of fruit load is obtained (Fig. 6). Fig. 6 326 suggests that R2 in the equations, and therefore the VPD influence on TGR, decreased 327 sharply from around 13 MT ha-1. 328 329 Discussion 330 MDS baseline usefulness 331 The relationship between the MDS signal and the midday stem water potential () was 332 similar to that described in the literature (olives, Moriana et al 2000; other fruit trees, 333 Ortuño et al 2010). When there was no water stress, the values for the relationship 334 between the MDS signal and the water potential (around -1.4MPa) were grouped around 335 1, while in the MDS vs.  relationship, these values showed a greater scattering. Such 336 results suggest that the MDS signal reduced the environmental noise which is common 337 in MDS values in the range near -1.4MPa. 338 The fruit load was a factor likely to affect the MDS signal vs.  relationship. 339 Conditions of full irrigation or very low water stress ( higher than -1.6MPa) in a low 340 fruit load season presented greater values than expected (Figure 4, higher than 1). The 341 fruit load is a factor that affects MDS values. In olive trees, Moriana et al (2011) 342 reported a significantly different lower slope in the baseline for the low fruit load than 343 for the high fruit load. Goldhamer and Fereres (2001) suggested that an active trunk 344 growth decreases the MDS in fruit trees. However, lower values for the MDS vs.  345 relationship were not found in low fruit load conditions in the present work (Figures 1 346 and 3). Since the MDS signal is a ratio, such response would be related to an estimation 347 of values lower than expected in conditions of full irrigation (the denominator in the 348 15 ratio). Therefore, the estimation of the MDS baseline at the beginning of a low fruit load 349 season, according to the Corell et al (2013) methodology, could underestimate the value 350 with full irrigation and then, produce a significant increase in the MDS signal during the 351 pit hardening. 352 The relationship between MDS signal and  showed a clear increase in the 353 MDS signal from -1.6MPa to -2MPa (Figure 4). Such increase was also observed in the 354 MDS vs.  relationship, although the variations were narrow and similar to some values 355 of the Control trees (Figure 3). In both relationships, values below -2MPa were similar 356 to the ones obtained with a higher than -1.4MPa. This pattern of increase and 357 decrease has been observed in olive (Moriana et al., 2000) and other fruit trees (Ortuño 358 et al., 2010) and has limited the usefulness of MDS in olive trees (Moriana and Fereres, 359 2002; Moriana et al 2003; Moriana et al., 2010; Fernández et al 2011). Although the 360 MDS signal also presented this pattern, MDS signal values greater than 1.1 always 361 indicated moderate water stress conditions. However, MDS signal values do not display 362 a linear increase because the decrease of MDS signal starts in this interval of water 363 potential. Therefore, in the interval 1.1-1.4, a higher MDS signal will not be necessarily 364 imply a lower Then, although there is still an uncertain zone in the range between -365 1.4MPa and -2MPa, at least conditions of water stress could be identified. 366 367 Relationship between trunk growth rate (TGR) and environment 368 TGR is poor related with environment in the literature and in the present work. 369 Predicted models of the daily TDF has reported no clear results for the overlap effect of 370 growth and water status (Deslauriers et al, 2007). Only in young olives trees, when 371 trunk growth is continuous during all the irrigation season because of the absence of 372 fruit development, significant relationships have been reported (Pérez-López et al., 373 16 2008; Cocozza et al., 2012). Deslauriers et al (2007) suggested in several species that 374 the relationship between TGR and temperature is strongly related with the rehydration 375 phase of the daily curve of trunk diameter variations. In the present work, no 376 relationships with any of the Deslauriers’ phases have been obtained. Fernández et al 377 (2011) in the same olive orchard did not obtain either any relationship. This lack of 378 results was likely related with the greater number of species and meteorological 379 conditions in the Deslauriers work than in Fernández and the present works. According 380 with the data of the present work, the influence of VPD was very important but strongly 381 affected for the yield. Both results are not new in olive literature. Evaporative demand 382 affects the daily cycle of leaf conductance (Xiloyannis et al., 1988) and the relationship 383 between leaf conductance and water potential (Moriana et al., 2002). Water relations are 384 strongly affected for fruit development (Rallo and Suárez, 1989; Martín-Vertedor et al., 385 2011). TGR in olive trees is very different in low than in high fruit load conditions 386 (Moriana et al., 2003), but, according to the present work, excessive fruit yield will also 387 affect. Moriana et al (2013) reported in two of the data set used in the present work 388 (2008 and 2010 seasons) a continuous decrease in the TGR values in full irrigated 389 conditions. Finally, the influence of VPD in TGR values was delayed in one day and the 390 increase in VPD affect the TGR of the next day. Such result suggests that TGR 391 variations could be controlled with chemical or hydraulic changes in the trunk tissues as 392 in the root signal, described also in olive trees (Fernández et al., 2006). 393 394 Conclusions 395 The patterns of the relationships MDS signal vs. and MDS vs.  were similar. 396 However, the MDS signal estimated according to Corell et al (2013) resulted in a 397 reduced scattering in conditions of full irrigation and clearly identified water stress 398 17 conditions in the range of -1.4MPa to -2MPa. This range of  corresponded to MDS 399 signal values between 1.1 and 1.4. However, since the decrease in MDS signal starts 400 within this range, higher values do not indicate more severe water stress conditions. 401 values lower than -2MPa produced values of MDS signal around 1, therefore, they 402 cannot be used for detecting water stress conditions. Conditions of low fruit load could 403 limit the usefulness of this approach. Significant relationship between TGR and 404 environmental variables were obtained only when a 1 day delayed was considered. TGR 405 values during pit hardening were strongly affected for the increase in the average VPD 406 of the day before when the fruit load was not excessive. 407 408 Acknowledge 409 This research was supported by the Spanish Ministerio de Ciencia e Innovación 410 (MICINN), (AGL2010-19201-CO4-03). Thanks are due to J. Rodriguez and A. 411 Montero for help with field measurements. 412 413 References 414 Allen, R.G., Pereira, L.S., Raes, D., Smith. M., 1998. Crop evaportranspiration. 415 Guideline for computing crop water requirements. FAO irrigation and drainage 416 paper nº 56. Roma. FAO 417 Cocozza, C., Giovannelli, A., Lasserre, B., Cantini, C., Lombardi, F., Tognetti, R. 2012. 418 A novel mathematical procedure to interpret the stem radius variation in olive 419 tres. Agric. Forest Meteorol. 161,80-93. 420 Corell, M., Girón, I.F., Moriana, A., Dell’Amico, J., Morales, D., Moreno, F. 2013. 421 Extrapolating base-line trunk shrinkage reference equations across olive 422 orchards. Agric. Water Manage. 126,1-8. 423 18 Cuevas, M.V., Torres-Ruiz, J,M., Álvarez, R,, Jiménez, M.D., Cuerva, J., Fernández, 424 J.E. 2010. Usefulness of trunk diameter variations for irrigation scheduling in a 425 mature olive tree orchard. Agric. Water Manage. 97, 1293–1302. 426 Deslauriers, A., Anfodillo, T., Rossi, S., Carraro, V. 2007. Using simple causal 427 modeling to understand how water and temperature affect daily stem variation in 428 trees. Tree Physiol. 27, 1125-1136. 429 Fernández, J.E. and Cuevas, M.V., 2010. Irrigation scheduling from stem diameter 430 variations: a review. Agric. Forest Meteorol. 150, 135–151. 431 Fernández, J.E., Moreno, F., Girón, I.F., Blázquez, O.M., 1997. Stomatal control of 432 water use in olive tree leaves. Plant Soil 190, 179-192. 433 Fernández, JE., Díaz-Espejo, A., Infante, J.M., Durán, P., Martín-Palomo, MJ., 434 Chamorro, V., Girón I.F., Villagarcía, L. 2006. Water relations and gas 435 exchange in olive trees under regulated deficit irrigation and partial rootzone 436 drying. Plant Soil 284, 273-291. 437 Fernández, J.E., Torres-Ruiz, J.M., Díaz-Espejo, A., Montero, A., Alvárez, R., Jiménez, 438 M.D., Cuerva, J., Cuevas, M.V. 2011a. Use of maximum trunk diameter 439 measurements to detect water stress in mature Arbequina olive trees under 440 deficit irrigation. Agric. Water Manage. 98, 1813-1821. 441 Fernádez, JE., Moreno, F., Martín-Palomo, MJ., Cuevas, MV., Torres-Ruiz, JM, 442 Moriana, A. 2011b. Combining sap flow and trunk diameter measurements to 443 assess water needs in mature olive orchards. Environ. Exp. Bot. 72, 330-338. 444 Goldhamer, D.A. and Fereres, E., 2001. Irrigation scheduling protocols using 445 continuously recorded trunk diameter measurements. Irrig. Sci. 20, 115-125. 446 19 Goldhamer, D.A., Fereres, E., Mata, M., Girona, J., Cohen, M., 1999. Sensitivity of 447 continuous and discrete plant and soil water status monitoring in peach tress 448 subjected to deficit irrigation. J. Amer. Soc. Hort. Sci. 124, 437-444. 449 Herzog, K.M., Hüsler, R., Thum, R., 1995. Diurnal changes in the radius of a subalpine 450 Norway spruce stem: their relation to the sap flow and their use to estimate 451 transpiration. Trees 10, 94-101. 452 Katerji, N., Tardieu, F., Bethenod, O., Quetin, P., 1994. Behavior of Maize stem 453 diameter during drying cycles: comparison of two methods for detecting water 454 stress. Crop Sci. 34, 165-169. 455 Klepper, B., Browing, V.D., Taylor, H.M., 1971. Stem diameter in relation to plant water 456 status. Plant Physiol. 48, 683-685. 457 Martín-Vertedor, A.I., Pérez-Rodríguez, J.M., Prieto, H., Fereres, E., 2011. Interactive 458 responses to water deficits and crop load in olive (Olea europaea L., cv. 459 Morisca).Water use, fruit and oil yield. Agric. Water Manage. 98, 941–949. 460 Moriana, A. and Fereres, E., 2002. Plant Indicators for Scheduling Irrigation for Young 461 Olive Trees. Irrig. Sci. 21, 83-90 462 Moriana, A., Fereres, E., Orgaz, F., Castro, J., Humanes, M.D., Pastor, M., 2000. The 463 relations between trunk diameter fluctuations and tree water status in olive tree 464 (Olea europea L). Acta Hortic. 537, 293-297. 465 Moriana, A., Orgaz, F., Fereres, E., Pastor, M., 2003. Yield responses of a mature olive 466 orchard to water deficits. J. Amer. Soc. Hort. Sci. 128,425-431. 467 Moriana, A., Girón, I., Martín-Palomo, M.J., Conejero, W., Ortuño, M.F., Torrecillas, 468 A., Moreno, F., 2010. New approach for olive trees irrigation scheduling using 469 trunk diameter sensors. Agric. Water Manage 97,1822-1828. 470 20 Moriana, A., Moreno, F., Girón, I., Conejero, W., Ortuño, M.F., Morales, D., Corell, 471 M., Torrecillas, A., 2011. Seasonal changes of maximum daily shrinkage 472 reference equations for irrigation scheduling in olive trees: influence of fruit 473 load. Agric. Water Manage. 99, 121-127. 474 Moriana, A., Pérez-López, D., Prieto, M.H., Ramírez-Santa-Pau, M., Pérez-Rodriguez, 475 J.M. 2012. Midday stem water potential as a useful tool for estimating irrigation 476 requirements in olive trees. Agric. Water Manage. 112:43-54. 477 Moriana, A., Corell, M., Girón, I.F., Conejero, W., Morales, D., Torrecillas, A., 478 Moreno, F. 2013. Regulated deficit irrigation based on threshold values of trunk 479 diameter fluctuation indicators in table olive trees. Sci Hort 164, 102-111. 480 Ortuño, M.F., Conejero, W., Moreno, F., Moriana, A., Intrigliolo, D.S., Biel, C., 481 Mellisho, C.D., Pérez-Pastor, A., Domingo, R., Ruiz-Sánchez, M.C., Casadesus, 482 J., Bonany, J., Torrecillas, A., 2010. Could trunk diameter sensors be used in 483 woody crops for irrigation scheduling?. A review of current knowledge and 484 future perspectives. Agric. Wat. Management 97,1-11. 485 Pérez-López, D.; Moriana, A.; Rapoport, H; Olmedilla, N.; Ribas, F. (2008) New 486 approach for using trunk growth rate and endocarp development in the irrigation 487 scheduling of young olive orchards. Scientia Hort. 155, 244-251 488 Pérez-López, D., Pérez-Rodríguez, J.M., Moreno, M.M., Prieto, M.H., Ramírez-Santa-489 Pau, M., Gijón, C., Guerrero, J., Moriana, A. 2013. Influence of different 490 cultivars-locations on maximum daily shrinkage indicators: Limits to the 491 reference baseline approach. Agric. Water Manage. 127, 31-39. 492 Rallo, L. and Suarez, M.P. 1989. Seasonal distribution of dry matter within the olive 493 fruit-bearing limb. Adv. Hort. Sci. 3, 55-59. 494 21 Rapoport, H.F., Pérez-López, D., Hammami, S.B.M., Aguera, J., Moriana, A. 2013. 495 Fruit pit hardening: physical measurements during olive growth. Ann. Appl. 496 Biol. 163, 200-208. 497 Scholander, P.F., Hammel, H.T., Bradstreest, E.A., Hemmingsen, E.A., 1965. Sap 498 pressure in vascular plant. Science 148, 339-346. 499 Xiloyannis,C; Pezzarossa,B; Jorba,J; Angelini,P (1988): Effects of soil water content on 500 gas exchange in olive trees. Adv. Hort. Sci. 2, 58-63. 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 Midday Stem Water Potential (MPa) -2.8 -2.6 -2.4 -2.2 -2.0 -1.8 -1.6 -1.4 -1.2 -1.0 -0.8 MDS Signal 0.6 0.8 1.0 1.2 1.4 1.6 1.8 Figure 2. Relationship between Midday stem water potential vs Maximum daily shrinkage signal (MDS signal) during the three seasons. Each symbol is the average of 6 measurements. The period of measurement was from the beginning of pit hardening until harvest. Symbols: 2011 season, triangles; up and empty Control trees, down and empty RDI 2, up and black RDI 12. 2012 season, square; empty Control trees, midfilled RDI 2; black RDI 12. 2013 season, circle; empty Control trees, mid filled RDI 2; black RDI 12. Vertical dash line indicated the reference value of stem water potential (- 1.4 MPa). Horizontal dash line indicated the reference value of MDS signal (1). Figure 2 Midday Stem Water Potential (MPa) -2.6 -2.4 -2.2 -2.0 -1.8 -1.6 -1.4 -1.2 -1.0 -0.8 MDS (µm) 300 400 500 600 700 800 900 Figure 3. Relationship between Midday stem water potential vs Maximum daily shrinkage during the three seasons. Each point is the average of all the data of Figure 1 grouped according to water potential intervals: values higher than -1.4 MPa, between - 1.4 until -1.75 MPa, between -1.75 until -2 MPa and lower than -2 MPa. Vertical and horizontal bars at the symbol represent the standard error in MDS and water potential respectively. Vertical dash line shows the reference of stem water potential (-1.4 MPa). Symbols: 2011 season, triangles; up and empty Control trees, down and empty RDI 2, up and black RDI 12. 2012 season, square; empty Control trees, mid-filled RDI 2; black RDI 12. 2013 season, circle; empty Control trees, mid filled RDI 2; black RDI 12. Figure 3 Midday Stem Water Potential (MPa) -2.6 -2.4 -2.2 -2.0 -1.8 -1.6 -1.4 -1.2 -1.0 -0.8 MDS SIgnal 0.6 0.8 1.0 1.2 1.4 1.6 Figure 4. Relationship between Midday stem water potential vs Maximum daily shrinkage signal (MDS signal) during the three seasons. Each point is the average of all the data of Figure 2 according to the water potential interval of: values higher than -1.4 MPa, between -1.4 until -1.75 MPa, between -1.75 until -2 MPa and lower than -2 MPa. Vertical and horizontal bars at the symbol represent the standard error in MDS signal and stem water potential respectively. Vertical dash line shows the reference of stem water potential (-1.4 MPa). Horizontal dash lines represent the reference of MDS signal (1) and an interval of ±10%. Symbols: 2011 season, triangles; up and empty Control trees, down and empty RDI 2, up and black RDI 12. 2012 season, square; empty Control trees, mid-filled RDI 2; black RDI 12. 2013 season, circle; empty Control trees, mid filled RDI 2; black RDI 12. Figure 4 (∆−1)VPD (KPa) -3 -2 -1 0 1 2 TGR (µm day -1 ) -300 -200 -100 0 100 200 300 TCT=-54.15(∆−1)VPD TCT=-79.39(∆−1)VPD Fig. 5. Relationship between trunk growth rate (TGR) and increment of the vapour pressure deficit the day before ((∆-1)VPD). Black square and solid line represent all the data of the 7*5 m orchard (Table 3, n=257, TGR=-54.15 (∆-1)VPD, R 2 =0.46***, Error=31.0 µm día - 1 ). White square and dash line represent data from 7*7 orchard (Table 3, n=60, TGR=- 79.39 (∆-1)VPD, R 2 =0.75***, Error=27.8 µm día -1 ) Figure 5 Yield (MTha -1 ) 6 8 10 12 14 16 18 20 Determination coefficient (R 2 ) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Fig. 6. Relationship between the determination coefficient (R 2 ) of the regressions between increment of the vapour pressure deficit the day before ((∆-1)VPD) and TGR (Table 3) vs the yield. The highest yield and lowest R 2 correspond to the regression obtained in the 2008 season (data not shown). Figure 6