Quantification of water uptake from pulsed input via disturbed and undisturbed channels on a desert bajada

Background/Question/Methods

In the Mojave Desert, during a heavy rain event, water rushes down a coalescence of alluvial fans and is distributed unevenly through naturally formed channels. When long corridors such as roads disrupt these natural flow paths, the majority of downstream channels receive little to no runoff water. Such disturbances cause unequal water distribution to vegetation down-slope. Plant proximity to channels has been shown to increase water uptake and plant production based on short-term measurements that require labor-intensive repeated sampling. Sap-flow gauges were used to provide automated and efficient long-term measurements after simulating rain events (pulses) at three sites on the foot of the Providence Mountains in the Mojave Desert: an upslope channel with natural flow (active channel), a channel below a road with interrupted flow (inactive channel), and an upslope area without a channel (simulated channel). For each of the three sites, nine plants (Larrea tridentata) located at different distances from margin of channel were instrumented. It was hypothesized that (1) in the active channel, plants located within 3m of the channel margin will obtain pulse water from within the channel, and (2) in inactive and simulated channels, only plants within 1m of the channel margin will obtain pulse water. 

Results/Conclusions

Plants within 3 meters of the active channel had a 20% sap-flow increase showing a maximum peak 16 days after pulse; whereas plants from inactive channel had a 45% sap-flow increase with a maximum peak 8 days after pulse. For plants in the simulated channel, only plants within 1 meter responded to pulse with a 20% sap-flow increase and a maximum peak 15 days after pulse. Plants located further than 3 meters from a channel did not respond to the pulses in any of the three sites. As opposed to expected results, plants located within 3 meters of the inactive channel did respond to the water pulse, and surprisingly showed higher percent increases in sap-flow, however, these values did not persist as in the case observed in the active and the simulated channels. It is not clear why plants near channels that have been cut off from significant flow for greater than 100 years would have a more pronounced pulse response, but the inability to sustain these higher rates may indicate a more surficial root distribution within the substrata influenced by the channel. Sap flow in response to natural rainfall pulses may help resolve these questions.