An open-source data-driven irrigation controller, "Open_Irr", was developed by USDA-ARS as a low-cost (≤ $250 USD) entry point to autonomous irrigation installations in horticultural production and research applications. The device is reliant on granular-matrix type Watermark soil tension sensors (n≤16) from which it acts as a datalogger; sensors can be partitioned into 4 groups for water management in which readings can be used to output a logic-level signal for opening valving to initiate irrigation events. Here, we demonstrate the use of the device in research applications for imposing drought stress in containerized plantings of apple (Malus domestica). Our primary objectives were to (i) determine the utility of using granular matrix type soil tension sensors for predicting plant physiologic responses, measured by growth habit, infrared gas exchange, and chlorophyll fluorescence, for improvement of decision support systems; and (ii) determine potential differences in cultivar performance under water deficit conditions. A trial examined performance of ‘B.9’ rootstock grafted to ‘Autumn Gala’, ‘CrimsonCrisp’, and ‘Golden Delicious’ scions at set soil matric potential thresholds for irrigation events (-25, -40, -60, and -80 kPa). A second trial examined five rootstock cultivars (‘B.10’, ‘B.9’, ‘G.11’, ‘G.935’, and ‘G.969’) grafted to a common ‘Honeycrisp’ scion as plant available water contents of the soil were theoretically progressively decreased by 5% each week. Correlation matrices between aspects of physiology and soil matric potential revealed Pearson’s r ≤ |0.43| yet further regression analysis offered potentially useful data shapes warranting further exploration. A piecewise regression suggested soil matric potential could predict, to an extent (r2 = 0.29), the rate of change in leaf water potential upon exposure to water deficit. The break-point of the piecewise regression in leaf and edaphic potentials was respectively -30.6 kPa and -0.73MPa, which may prove useful for incorporation in developing irrigation decision support systems. The pattern of stomatal response to deficit suggested enhanced stomatal regulation capacity of ‘Gala’ relative to other scions in which nonstomatal factors likely contributed to short term maintenance of photosynthetic C assimilation rates. Similarly, the ‘G.935’ rootstock demonstrated superior sensitivity to water deficit through rapid stomatal closure relative to other cultivars. Refinement of the relationships between arduous measures of physiologic function with correlated easy-to-automate proxy measures is advised for potential advancement of irrigation decision support systems.
