HYDRAULIC REDISTRIBUTION BY ROOTS: A Mechanism of Interaction between the Biosphere Moisture Reservoirs of the Deep , the Near-Surface Soil, and the Lower Atmosphere Geremew G. Amenu and Praveen Kumar Civil & Environmental Engineering, University of Illinois at Urbana-Champaign, IL 61801,USA

θ 3. MODEL VAPOR FLUX Demand 1. INTRODUCTION Flow in the MODULE PRINCIPALψ ALGORITHM root Based on Darcy’sρ Law ψ Objective: INPUT ∂ ∂ρ ⎡ ⎛ ∂ soil ⎞⎤ Flow in the Hydraulic Soil Model − Ksoil ⎜ ψ +1⎟ = − root Kr,avg ( soil −ψ xylem ) SOURCE Xylem ∂t ∂z ⎢ ∂z ⎥ To model hydraulic redistribution and investigate its effect on (a) DATA soil pores Redistribution ⎣ ⎝ ⎠⎦ Based on Hagen-Poiseuille’sρ Law profile and (b) fluxes at land-atmosphere interface in water-limited . ⎡ ∂ ⎤ ψ Atmos. ∂ ⎛ xylem ⎞ Root Model − ⎢ root Kroot ⎜ +1⎟⎥ = root Kr,avg ( soil −ψ xylem ) NARR Soil ∂z ⎣ ⎝ ∂z ⎠⎦

2 m Forcing Motivation: ≈ Layer Flow between Phloem - The long-memory nature of deep soil layer Soil Global Soil soil and root Stomatal Jarvis Algorithm Conductance gs = gs,max f (Rad) f (T ) f (ψ leaf ) f (D) f (C) - The availability of plant roots at deeper soil layers 5 m Properties Data Task ≈ SOIL Penman-Monteith Algorithm MEDIA ROOT Canopy [∆.Q + ρ c (e − e )g ]g Hypothesis: Land Cover IGBP-DIS LIQUID FLUX MEDIA E = λn a p s a a c Supply Transpiration [∆.g + γ (g + g )] “The hydrologically active depth of rooting zone in current climate models is c c a Vegetation underestimated and that the long-memory deep soil-layer may have significant AVHRR Fig.2: Conceptual representation of the soil-plant-atmosphere continuum (left) and the Carbon Farquhar-Collatz Algorithm Properties A = min(A , A , A ) impact on land-atmosphere interaction in vegetated environments.” hydraulic redistribution model (right) as used in this study. Assimilation c q e

DAILY MOISTURE UPTAKE (mm/s) PROFILE -6 UPTAKE PROFILE DURING DAY & NIGHT HOURS -8 x 10 x 10 2. STUDY SITE 4. RESULTS 6 300 2 2 6 Without HR With HR 200 Effect of HR on Difference between Moisture Profiles of Simulations with and without Hydraulic Redistribution (HR) 4 Transpiration 4 4 Mid-Night 4 2 100 0.05 4 2 0 6 6 6 2 0 8 -0.05 0 Mid-Day 10 -0.1 0 (%) trans. in Change -100 12 8 8 (m/s) rate Transpiration 2 4 6 8 10 12 1 2 3 4 5 6 7 8 9 10 11 12 Soil Layer Index 0 24 48 72 96 120 144 168 192 216 240 264 288 312 Soil Layer Index -2 Time (month) Time (month) 180 Time (months) 160 10 10 s) Boundary of 140 2 14 300 -4 study site 120 12 Without HR With HR 1 1 1 Effect of HR on 100 12 12 mol/m 200 µ Assimilation 80 2 2 2 5 10 15 20 -1 -0.5 0 0.5 1 9 Fresno 60 -5 Time (hrs) Uptake (mm/s) x 10 Precipitation (mm) Precipitation 40 100 3 3 3 6 20 SEASONAL MOISTURE UPTAKE (mm/s) PROFILE x 10-6 UPTAKE PROFILE DURING WET & DRY MONTHS Soil Index Layer 0 4 4 4 0 n ar t 3 Ja Jul ug Feb M Apr May Jun A Sep Oc Nov Dec 5 5 5 2 10 2 Change in assim. (%) assim. in Change 0 -100 Fig.1: Location of case study site – Sierra Nevada 6 6 6 8 rate ( Assimilation 2 4 6 8 10 12 1 2 3 4 5 6 7 8 9 10 11 12 Winter The 4 4 Time (month) Time (month) 7 7 7 6 (DJF) Whole 8 8 8 Period 4 Root Fraction (%) Summer Wet Season 2 400 6 6 (Jan) Soil Depth & Root Distribution 9 9 9 Dry Season Without HR 0 5 10 15 20 25 (JJA) 2 Effect of HR (Jul) With HR 300 0.01 1.5 on WUE 10 10 10 0 The depth to center of ith soil layer: 8 8 200 Soil Layer Index 11 11 11 Soil Layer Index -2 1 0.5(i−0.5) 100 zi = 0.025 (e −1) 0.1 12 12 12 10 -4 10 0.5 0 th The root fraction in the i soil layer: -50 -25 0 25 50 -50 -25 0 25 50 75 90 -50 -25 0 25 50 -6 0 Change in WUE (%) -100 −2 12 12 c−1 ⎡ c ⎤ 1 2 4 6 8 10 12 -1 -0.5 0 0.5 1 1.5 2 4 6 8 10 12 1 2 3 4 5 6 7 8 9 10 11 12 Soil depth (m) Soil depth c ⎛ z ⎞ ⎛ z ⎞ 1.27875 (WUE) Water-use-efficiency i i Change in Moisture (%) -5 Time (month) Time (month) Froot (zi ) = −∆z ⎜ ⎟ ⎢1+ ⎜ ⎟ ⎥ , c = Time (months) Uptake (mm/s) x 10 z ⎜ z ⎟ ⎜ z ⎟ ⎛ z ⎞ 50 ⎝ 50 ⎠ ⎣⎢ ⎝ 50 ⎠ ⎦⎥ log⎜ 50 ⎟ ⎜ z ⎟ Fig.4: The effect of Hydraulic Redistribution (HR) on soil moisture profile. Positive Fig.5: Daily (top) and Seasonal (bottom) profiles of water ⎝ 95 ⎠ 10 Fig.6: The effect of HR on Transpiration (top), Photosynthesis Where z and z are vegetation-dependent constants. (negative) values indicate an increase (decrease) in soil moisture due to HR. uptake by vegetation root system. 50 95 Fig.3: Root Profile (middle) , & Water-Use-Efficiency (bottom).

5. CONCLUSIONS 6. ACKNOWLEDGEMENTStridentata Hydraulic redistribution (HR) – a passive transport of soil water across soil layers via plant roots – could be modeled by Support for the research is provided by the National Oceanic and Atmospheric Administration (NOAA) Grant NA03OAR4310070. Computational resources for the project is partly provided by the National Center for Supercomputing Applications (NCSA). assuming the plant root system as a conduit for moisture transport along a pressure gradient, analogous to pipe flow. HR coupled with rooting depth could be a main mechanism for facilitating the dynamic interactions between moisture 7. REFERENCES reservoirs of the long-memory deep soil, the medium-memory near-surface soil, and the short-memory atmosphere. 1. Amenu, G.G., P. Kumar, and X.-Z. Liang, 2005: Interannual Variability of Deep-Layer Hydrologic Memory and Mechanisms of Its Influence on Surface Energy Fluxes. J. Climate, 18, 5024-5045. The incorporation of HR results in a significant (i) alteration in the profiles of soil moisture and moisture uptake, and (ii) 2. Canadell et al., 1996: Maximum Rooting Depth of Vegetation Types at the Global Scale. Oecologia, 108,583-595. increase in dry-season transpiration, carbon assimilation, and water-use-efficiency. 3. Farquhar, et al., 1980: A Biochemical Model of Photosynthetic CO2 Assimilation in Leaves of C3 . Planta, 149, 78-90. 4. Jarvis, P.G., 1976: The Interpretation of the variations in leaf and stomatal conductance found in canopies in the field. By enhancing the effectiveness of water uptake by plant roots from long-memory deep soil moisture reservoirs (and hence, Phil. Trans. R. Soc. Lond., B. 273, 593-610. 5. Richards, J.H., and M.M. Caldwell, 1987: Hydraulic Lift: Substantial Nocturnal Water Transport between Soil Layers by Artemisia controlling the water, energy, and carbon cycles), the dynamic mechanism of HR may have an important implication for Roots. Oecologia, 73, 486-489. understanding processes that govern long-term climate and ecological predictions. 6. Schenk, H.J., and R.B. Jackson, 2002: The Global Biogeography of Roots. Ecological Monographs, 72, 311-328.