No abstract

To improve the scheduling of irrigation for low-density olive trees (Olea europaea L.) grown in a typical Mediterranean environment of Southern Portugal, and to clarify the mechanisms of water uptake by trees, transpiration, soil water status and stomatal response to water deficit were measured in an olive orchard. Olive trees of cv. Cordovil were subject to three irrigation treatments: full-rate irrigation, sustained deficit irrigation (SDI) providing for approximately 60% of water applied at full-rate irrigation, and a regulated deficit irrigation (RDI) with water applied at periods during three critical phases: before-flowering, at beginning of pit-hardening, before crop-harvesting to replenish soil moisture to field capacity. There was also a dry-farming treatment. Trees responded differently to summer rainfall and irrigation water: full-rate irrigation, which received 880 mm of irrigation and 240 mm of rainfall, used 704 mm for transpiration; SDI, which received the same amount of rainfall and 448 mm of irrigation water, used 745 mm of water for transpiration; RDI, which received 69 mm of irrigation water and 240 mm of rainfall, used 638 mm of water for tree transpiration; dry-farming, which received no irrigation, benefited from 240 mm of summer and early autumn rain and used 404 mm of water for transpiration. The results support the hypothesis that trees under RDI and dry-farming satisfy most of their early atmospheric evaporative demand by extracting water from outside of the area wetted by drip irrigation. Scaled-up orchard transpiration was used to define orchard crop and water stress coefficients. With full-rate irrigation and SDI the results showed that during summer droughts olive trees slow down their physiological mechanisms to conserve water, regardless of amount applied. The derived crop coefficient results also indicated that SDI was the most appropriate for scheduling the irrigation of cv. Cordovil orchards in Southern Portugal although applying RDI helped sustain orchard transpiration and yields. Irrigation accounted for 11% of total water used in transpiration, with the balance extracted by roots in the large volume of soil lying in the areas between the trees. However, using the RDI scheme to schedule irrigation appears to be appropriate only in wet years with well distributed late summer rainfall or where there is a shortage of farm irrigation water. In general, and particularly in years with no summer and early autumn rains as can often occur in this region, the SDI regime appears to be more appropriate for scheduling irrigation.

Experimental results obtained in Southern Portugal from a dry-farmed mature olive tree orchard recently converted to drip irrigation are described. Water use and response to two irrigation management practices by olive trees was monitored with sap flow compensation heat pulse sensors, 'Watermark' granular matrix block sensors and a capacitance probe.The 80-plus-year-old mature olive tree orchard planted on a 12 m by 12 m spacing layout was converted in 2005 from dry-farming to drip irrigation and subjected to two water treatments: trees irrigated daily to supply for crop water demand and trees irrigated beforeflowering, during pit-hardening and before crop-harvesting. Sap flow sensors were implanted in sample trees at three different positions around the trunk and measurements were taken at 30 min intervals during 4 months, from April to mid-August of 2005. Tree transpiration rates were estimated as average of sap flow rates. When trees were fully irrigated, the observed differences in daily sap flow rate amplitude were explained by the natural trees difference in canopy cover, plant height and conductance of water vapour sites. However, when deficit irrigation was prescribed and, when the trees stopped being irrigated, they gradually lost their ability to adequately respond to the evaporative demands of the day, showing smaller variations in amplitudes sap flow. After irrigation ceased in May 15, transpiration rate gradually decreased from its maximum of 7 l h À1 , when trees were fully irrigated and soil water content was near to field capacity, to values of less than 3 l h À1 by July 3 as the soil water content gradually acted as the transpiration limiting factor.Transpiration rates recovered after irrigation was re-introduced on July 4. Although low in the non-irrigation period, transpiration rates never dropped to zero and stayed between 37 and 50 l d À1 from May 27 to June 9, as trees were able to extract soil water in the absence of irrigation. Olive trees maintained transpiration to levels as high as 50 l d À1 suggesting that long after irrigation is suppressed, a considerable amount of water held in the soil is made available to the trees. Differences in evapotranspiration and transpiration rates during the same period also indicated that olive trees, making use of the extensive root system developed in the 12 m by 12 m tree spacing, were able to extract soil water and maintain transpiration levels as high as 50 l d À1 , while soil water balance indicated tree evapotranspiration rates close to zero. This particular ability of dry-farmed olive trees to remove water held in the soil under adverse conditions of very low soil moisture and uncertainties associated with the real volume of soil effectively explored by the root system, make profile probe sensors, regardless of their accuracy, unsuitable for control of water uptake and management of dry-farmed olive orchards recently converted to irrigation. Likewise, ARTICLE IN PRESS

Tree water uptake is often estimated based in a crop coefficient k c , a ratio of the tree water uptake and a reference evapotranspiration, ET0. The concept behind estimating tree crop coefficient implies that data should be representative of the population of trees analyzed. Ideally it would require the monitoring of a large number of trees in each treatment population. This paper reports on a scaling method to establish stand-level transpiration estimates and crop coefficients from individual sampled tree sap flow measurements. The scaling technique was implemented for individual tree sap flow measurements on the following irrigation treatments: A, fully-irrigated; B, irrigated to provide for approximately 60% of crop evapotranspiration; C, irrigated to provide for 100% of crop evapotranspiration during three critical phase periods: before-flowering, at beginning of pit-hardening and before crop-harvesting, and dry-farming treatment D. Results show that stand transpiration T depart from individual tree transpiration values. They consequently were used to establish crop, k c and water stress, k s coefficients to account for the cluster's characteristics and degree of tree's water uptake. Using the individual tree transpiration rates would be less appropriate.

No abstract