Background and aims Water is often heterogeneously distributed in soils. Understanding roots' responses to this soil-water heterogeneity is a key parameter defining plant survival in dry climates. To determine local root water uptake for partly dry conditions in a plant's root system, we prepared soil patches with different water contents, then used neutron radiography to monitor daily changes in root structure and water uptake. Methods Lupin plants were grown in 30×25×1 cm 3 aluminum containers filled with sandy soil. In two partitioning set-ups, the soil-root zone was divided into either two or nine hydraulically-isolated soil compartments. This was done by packing layers of coarse sand as capillary barriers, by which vertical and/or horizontal soil water heterogeneity as well as homogeneous wellwatered treatments were applied for control. Daily changes in soil water content in each compartment, water uptake and root growth were monitored noninvasively and quantified by neutron radiography during a period of 15 consecutive days. Results In optimal homogeneously-wet soil, lateral roots in the top 10 cm of the root system showed the highest water uptake rate, up to around 10 mg/(mm. root. day), which on average was twice as much as that for younger lateral roots in lower position and taproot. In heterogeneous treatments, root water uptake declined strongly in compartments with the soil water content below 0.13-0.10 cm 3 /cm 3 while in parallel an enhanced uptake rate, rising by up to 100 %, was observed for the roots in wet compartments, presumably to compensate for roots in dry compartments and, therefore, sustain the total transpiration. Also, our observations showed that in the drying compartment a reduction of soil water content to 0.10-0.15 triggered local cluster root formation. Conclusions With the experimental set-up presented the pattern of water uptake across a lupin root system can be quantified and normalized to root length. Water uptake was shown to be highly variable in different parts of the root system. A threshold for water stress to cause cessation of local water uptake was identified, and the considerable amount of compensation by water uptake in other parts identified. The dynamic trade-off among different parts of the root system seems to regulate total root uptake also during water stress to sustain the daily transpirational demand. Plant Soil

Cover crops have been shown to deplete soil nitrate and reduce nitrate leaching in Mediterranean climates. Cover crop canopy development, N uptake, and root system development were studied during three rainy seasons in Yolo silt loam and Rincon silty clay loam soils, with two cover crops with contrasting root systems and N acquisition strategies: triticale (× Triticosecale), a monocot wheat (Triticum aestivum L.)–rye (Secale cereale L.) hybrid with a fibrous root system, and bell bean (Vicia faba L.), a taprooted legume that fixes N. Weed growth was substantial in bell bean plots. Triticale consistently developed a uniform stand and produced a deeper and denser root system with a faster root penetration rate than the bell bean–weed mixture. Growth and N uptake were strongly correlated with accumulated temperature for triticale but not bell bean. Therefore, triticale growth models based on weather data (1982–2014) were used to study the effect of planting and termination dates on N uptake in Davis, CA. Planting before mid‐October would enable N extraction from the soil prior to the heaviest winter rains, whereas delaying sowing until mid‐November would delay significant N uptake until early spring. Following the current cover crop management approach of sowing triticale before the first significant rain and terminating by the end of February, N uptake from the soil would exceed 45 kg N ha‐1 ∼56% of the time. Generally, a temperature‐based growth model is useful for managing soil nitrate levels with some monocot cover crops during the winter rainy season.

The conventional tensiometer is among the most accurate devices for soil water matric potential measurements, as well as for estimations of soil water flux from soil water potential gradients. Uncertainties associated with conventional tensiometers such as caused by ambient temperature effects and the draining of the tensiometer tube, as well as their limitation for deep soil monitoring, has prevented their widespread use for vadose zone monitoring, despite their superior accuracy in general. We introduce an alternative tensiometer design that offers the accuracy of the conventional tensiometer, while minimizing the aforementioned uncertainties and limitations. The proposed alternative tensiometer largely eliminates temperature-induced diurnal fluctuations and uncertainties associated with draining of the tensiometer tube and removes the limitation in installation depth. In addition, the manufacturing costs of this alternative tensiometer design are close to those of the conventional tensiometer, while it is especially suited for monitoring of soil water potential gradients as required for soil water flux measurements.