Mean annual infiltration above the high-level waste repository proposed to
be sited at Yucca Mountain, Nevada, has a large impact on assessments of re
pository performance. Ongoing investigations of infiltration processes have
identified the relatively horizontal caprock environment above portions of
the repository as a potentially large source of infiltrating waters, due t
o shallow, permeable soils above a moderately welded tuff with large soil-f
illed fissures. The combination of shallow soils and fissured bedrock allow
s rapid penetration of wetting pulses to below the rooting zone. Plant upta
ke can strongly reduce net infiltration in arid environments with high wate
r storage capacity, and, despite the low water storage capacity, there is a
relatively high vegetation density in this environment. The apparent discr
epancy between high vegetation density and low water storage motivates the
study of plant-hydrologic interactions in this semiarid environment. Field
observations were coupled with plant- and landscape-scale models to provide
insight into plant-hydrologic interactions. Several lines of evidence, inc
luding: (i) linear plant growth features observed on aerial photographs; (i
i) comparisons of plant cover within the fissured environment and comparabl
e environments lacking fissures; and (iii) direct excavations, all suggest
that the widely spaced soil-filled fissures are conducive to plant growth e
ven when fissures are buried at soil depths exceeding 30 cm. Results from a
mechanistic simulation model for root growth into fissures suggest that th
e additional (sheltered) plant-available soil water within fissures provide
s a competitive advantage for plant establishment. Therefore, plants that g
erminate above a fissure are more likely to survive, in turn developing lin
ear features above fissures. Having established that plants preferentially
root within soil-filled fissures in the caprock environment, a set of simul
ations were performed to examine the hydrologic consequence of plant roots
within fissures at the landscape-scale. The response to three rainfall amou
nts was simulated. For the largest storm, fluxes at the fissure bottom peak
ed at 1-4 weeks after the storm when plant uptake was not active, but were
eliminated when fissures had active vegetation. When plants were active wit
hin a fissure, uptake eliminated net infiltration in the fissure regardless
of the size of the storm. Two plant-related mechanisms reduced total flux
through the plant-filled fissures: (i) transpiration during fissure flow, a
nd (ii) wetting-pulse retardation due to drier fissures prior to rain. The
first mechanism appears to be dominant in these simulations. Results sugges
t that transpiration may strongly limit net infiltration (i.e. total deep p
ercolation flux escaping the plant root zone); significant infiltration can
occur, however, when plants are dormant, so that most infiltration would b
e expected to occur during winter. (C) 1999 Elsevier Science B.V. All right
s reserved.