The amelioration of plant salinity tolerance due to reduction in potential evapotranspiration is a long recognized phenomenon. In spite of this, salinity tolerance of plants is generally calculated from full season, time-and space-averaged response data. We hypothesized that the HYDRUS-1D model could be used to predict dynamic changes in plant salinity tolerance for a greenhouse vegetable crop over a full season and to determine best management practices regarding blending of saline with desalinated water for optimization of yields and water use efficiency (WUE). The specific objectives of the study were to determine dynamic vapor pressure deficit (VDP)-salinity response relationships of bell pepper plants grown in lysimeters and to apply them for hypothetical management scenarios when irrigating with blended desalinated and brackish water under commercial conditions. The transpiration response of bell pepper plants to salinity in the controlled lysimeter experiment was strongly influenced by variations in potential transpiration throughout the season. The plants were relatively tolerant during periods of low VPD and relatively sensitive during periods of high transpiration demand. Data were used to develop salinity response equations as a function of VPD. In a case study for Israel's Arava Valley, transpiration and water productivity of bell peppers could be increased 5% by blending saline and desalinated water such that less saline water was applied during periods of relatively high sensitivity (high VPD) and more during periods of relative tolerance as compared to application of the same total of both sources of water blended at a constant ratio throughout the season. Sensitivity analysis of the dynamic crop response model revealed that such increases in water productivity would be even greater for more salt sensitive crops.