Root length density and soil water distribution in drip-irrigated olive orchards in Argentina under arid conditions

Abstract

Several studies have evaluated many above-ground aspects of olive production, but essential root system characteristics have been little examined. The objective of our study was to evaluate root length density (RLD) and root distribution relative to soil water content in three commercial orchards (north-west Argentina). Depending on the orchard, the different drip emitter arrangements included either: (1) emitters spaced continuously at 1-m intervals along the drip line (CE-4; 4 emitters per tree); (2) 4 emitters per tree spaced at 1-m intervals, but with a space of 2m between emitters of neighbouring trees (E-4); or (3) 2 emitters per tree with 4m between emitters of neighbouring trees (E-2). All of the orchards included either var. Manzanilla fina or Manzanilla reina trees (5-8 years old) growing in sandy soils, although the specific characteristics of each orchard differed. Root length density values (2.5-3.5cm/cm3) in the upper soil depth (0-0.5m) were fairly uniform along the drip line in the continuous emitter (CE-4) orchard. In contrast, roots were more concentrated in the E-4 and E-2 orchards, in some cases with maximum RLD values of up to 7cm/cm3. Approximately 70% of the root system was located in the upper 0.5m of soil depth, and most of the roots were within 0.5m of the drip line. For each of the three orchards, significant linear relationships between soil water content and RLD were detected based on 42 sampling positions that included various distances from the trunk and soil depths. Values of RLD averaged over the entire rooting zone and total tree root length per leaf area for the three orchards were estimated to range from 0.19 to 0.48cm/cm3 and from 1.8 to 3.5km/m2, respectively. These results should reduce the uncertainty associated with the magnitude of RLD values under drip irrigation as intensively managed olive orchards continue to expand in established and new growing regions.