Abstract The effects of deficit irrigation applied during different phenological periods, and of different crop load levels on the water relations and on tree productivity of young Japanese plum cv. Black-Gold trees have been evaluated. Moreover, the usefulness of WatermarkŪ sensor for measuring matric soil water potential (?m) and of stem diameter sensor to estimate plant water status has been investigated. This mid-season maturing cultivar showed different sensitivity to water deprivation depending upon the phenological period. Thus, deficit irrigation applied during post-harvest did not affect flowering, fruit set or fruit growth the next season, allowing for a 29% water saving. However, several seasons of drought after harvest could reduce productivity of young trees as a consequence of the cumulative effect of water deficit on tree growth. Nonetheless, the smaller tree size, allows easier pruning and fruit thinning, with potential cost-saving for producers. On the other hand, water restriction applied during stages II and III of fruit growth, reduced fruit size in proportion to the integrated water stress achieved and also affected tree growth when water restrictions were either severe or in conjuction with a high crop load level. Severe water deficit applied only during stage II of fruit growth and the beginning of stage III, reduced fruit size by a 12% but advanced fruit maturation probably due to the important reduction of shoot growth. Drouhgting based on a moderate, but continuous reduction of irrigation rates from phase II of fruit growth until the end of the season, led to a too severe reduction of tree growth, and may also reduce fruit size if integrated water stress during fruit growth overcomes the 45 MPa·day threshold value. In a one-year experiment, increasing crop load from 4-5 fruit/cm2 of trunk cross sectional area (TCSA) to 7-8 fruit/cm2 of TCSA, allowed for a 47% increase in yield, as average fruit weight was only reduced by a 16%. High crop load also reduced the vegetative growth, with higher effect on trunk growth than on canopy development. It enhanced the sensitivity to water stress of trunk growth but not that of fruit growth. Crop load also affected tree water relations. In well irrigated trees, high crop load considerably increased the maximum diurnal trunk shrinkage (MDS) and reduced daily trunk growth (TGR). Thus, these feactures have to be taken in consideration when using the short term trunk diameter variations for irrigation scheduling. MDS, TGR and the water potential of non-transpiring leaves (?stem) all responded well in agreement to water reductions, therefore they are useful water stress indicators. However, higher variability of readings was obtained in MDS and TGR that on ?stem, and therefore ?stem was the most sensitive indicator. Reference, base-line equations, have been obtained to predict ?stem and MDS values as a function of the evaporative demand. These reference equations can be a useful guidelines for scheduling irrigation protocols based on ?stem and MDS measurements. A decrease in plant water status towards the end of the season occurred even in the well irrigated trees. Thus, for ?stem, different reference equations have to be used at the beginning of the season (fruit growth) than at the end (after harvest). A seasonal change in the relation between MDS and ?stem was observed, which compensated the decrease in plant water status, and therefore similar MDS values for well irrigated trees can be expected before or after harvest. The influence of tree size on the relationship between MDS and ?stem was also investigated. In a range of tree trunk diameter between 8 to 14 cm, MDS increased a 13% for each cm of increase of trunk diameter, due to the thicker phloem tissues of larger trees. This result may allow to extrapolate ?stem prediction based on empirical relations with MDS to plum trees of different sizes. WatermarkŪ sensors readings were in reasonable agreement with the irrigation regime particularly under severe water restrictions. They showed good indication of plant water status in the dry range of ?m, but not in the wet range (-20 kPa