Impacts of Root Hydraulic Redistribu5on on Site, Regional, - - PowerPoint PPT Presentation

impacts of root hydraulic redistribu5on on site regional
SMART_READER_LITE
LIVE PREVIEW

Impacts of Root Hydraulic Redistribu5on on Site, Regional, - - PowerPoint PPT Presentation

Impacts of Root Hydraulic Redistribu5on on Site, Regional, and Global Evapotranspira5on and Soil Moisture W.J. Riley Jinyun Tang, Jie Niu Earth


slide-1
SLIDE 1

Impacts ¡of ¡Root ¡Hydraulic ¡Redistribu5on ¡

  • n ¡Site, ¡Regional, ¡and ¡Global ¡

Evapotranspira5on ¡and ¡Soil ¡Moisture ¡

W.J. ¡Riley ¡ ¡ Jinyun ¡Tang, ¡Jie ¡Niu ¡ ¡ Earth ¡Sciences ¡Division ¡ Lawrence ¡Berkeley ¡Na5onal ¡Laboratory ¡

slide-2
SLIDE 2

Root ¡Hydraulic ¡Redistribu5on ¡

  • Basic ¡concept ¡
  • Lee ¡et ¡al. ¡(2005), ¡Wang ¡(2011) ¡

– RHD ¡increases ¡ET ¡and ¡photosynthesis ¡during ¡dry ¡ season ¡(but ¡of ¡intermediate ¡soil ¡wetness ¡as ¡in ¡Wang?) ¡ – Enhances ¡ground ¡water ¡deple5on/recharge. ¡ ¡ – Modifies ¡seasonal ¡climate ¡

  • Amenu-­‑Kumar ¡(2008) ¡
  • ­‑ ¡Deep ¡roots ¡RHD ¡enhances ¡the ¡connec5on ¡

between ¡surface ¡and ¡ground ¡water ¡

Amenu ¡and ¡Kumar ¡(2008) ¡ ¡

slide-3
SLIDE 3

Background ¡

  • Lee ¡et ¡al. ¡(2005), ¡Wang ¡(2011) ¡

– Root ¡Hydraulic ¡Redistribu5on ¡(RHD) ¡increases ¡ET ¡ and ¡photosynthesis ¡during ¡dry ¡season ¡ ¡ – Enhances ¡ground ¡water ¡deple5on ¡and ¡recharge ¡ – Modifies ¡regional ¡climate ¡

  • Amenu ¡and ¡Kumar ¡(2008) ¡

– Deep ¡root ¡hydraulic ¡redistribu5on ¡enhances ¡the ¡ connec5on ¡between ¡surface ¡and ¡ground ¡water ¡

slide-4
SLIDE 4

Objec5ves ¡

  • Evaluate ¡impacts ¡of ¡RHD ¡on ¡hydrological ¡

states ¡and ¡fluxes ¡

– Implement ¡RHD ¡in ¡CLM4.5 ¡using ¡the ¡Amenu-­‑ Kumar ¡model ¡ – Test ¡impacts ¡of ¡numerical ¡implementa5on ¡ – Test ¡impacts ¡of ¡roo5ng ¡distribu5on, ¡depth, ¡and ¡ proper5es ¡ – Test ¡impacts ¡of ¡pedotransfer ¡func5on, ¡including ¡ for ¡oxisols ¡

8/13/14 ¡ 4 ¡

slide-5
SLIDE 5

The ¡Amenu-­‑Kumar ¡(2008) ¡Model ¡of ¡RHD ¡

5 ¡

slide-6
SLIDE 6

Sequen5al ¡Coupling ¡vs. ¡Tight ¡Coupling ¡

Sequen&al ¡model ¡

  • Process-­‑spli_ng ¡method ¡

Tightly ¡coupled ¡model ¡

  • Form ¡and ¡solve ¡

coupled ¡system ¡

6 ¡

∂θ ∂t = ∂ ∂z Ksh ∂ψsm ∂z −1 # $ % & ' ( ) * + ,

  • .− Krh,rad ψsm −ψrp

( )

0 = ∂ ∂z Krh,ax ∂ψrp ∂z −1 # $ % & ' ( ) * + ,

  • .+ Krh,rad ψsm −ψrp

( )

Step ¡1: ¡solve ¡Richards’ ¡equa5on ¡ Step ¡2: ¡solve ¡root ¡model ¡

slide-7
SLIDE 7

1 2 3 4 5 6 7 1 2 3 4 5 6 7 ET (mm day−1) from CLM4.5RHD−SM with ∆t=1 min ET (mm day−1) (a) 1 2 3 4 5 6 7 1 2 3 4 5 6 7 ET (mm day−1) from CLM4.5RHD−TM with ∆t=30 min ET (mm day−1) (b) CLM4.5RHD−SM ∆t=10 min vs. CLM4.5RHD−SM ∆t=1 min CLM4.5RHD−SM ∆t=30 min vs. CLM4.5RHD−SM ∆t=1 min CLM4.5RHD−TM ∆t=1 min vs. CLM4.5RHD−TM ∆t=30 min CLM4.5RHD−TM ∆t=10 min vs. CLM4.5RHD−TM ∆t=30 min CLM4.5RHD−SM ∆t=1 min vs. CLM4.5RHD−TM ∆t=30 min

Sequen5al ¡Model ¡(SM) ¡Has ¡Large ¡ Sensi5vity ¡to ¡Time ¡Step ¡

7 ¡

SM ¡model ¡bias ¡

slide-8
SLIDE 8

−1.5 −1 −0.5 0.5 1 1.5

Sequen5al ¡Coupling ¡vs. ¡Tight ¡Coupling: ¡ Non-­‑Physical ¡Changes ¡in ¡Global ¡ET ¡

8 ¡

Coupling ¡biases ¡are ¡as ¡high ¡as ¡3.5 ¡mm ¡day-­‑1 ¡

(b) JJA

(mm ¡d-­‑1) ¡

slide-9
SLIDE 9

Blodgee ¡Forest ¡Site ¡

hep://journalism.berkeley.edu/ ¡ hep://fluxnet.ornl.gov/ ¡

slide-10
SLIDE 10

Sierra ¡(Blodgee) ¡LH ¡Evalua5on ¡

10 ¡

0.5 1 1.5 2 2.5 3 3.5 CLM4.5 Month Fluxnet MTE ET ( mm/day ) 1 2 3 4 5 6 7 8 9 10 11 12

slide-11
SLIDE 11

0.5 1 1.5 2 2.5 3 3.5 CLM4.5

  • Month

Fluxnet MTE ET ( mm/day ) 1 2 3 4 5 6 7 8 9 10 11 12

Sierra ¡(Blodgee) ¡LH ¡Evalua5on ¡

11 ¡

slide-12
SLIDE 12

0.5 1 1.5 2 2.5 3 3.5 CLM4.5

  • Month

ET ( mm/day ) Fluxnet MTE 1 2 3 4 5 6 7 8 9 10 11 12

Sierra ¡(Blodgee) ¡LH ¡Evalua5on ¡

12 ¡

slide-13
SLIDE 13

0.5 1 1.5 2 2.5 3 3.5 CLM4.5

  • Month

ET ( mm/day ) Fluxnet MTE 1 2 3 4 5 6 7 8 9 10 11 12

Sierra ¡(Blodgee) ¡LH ¡Evalua5on ¡

13 ¡

slide-14
SLIDE 14

20 40 60

  • Prec. (mm day−1)

Ordinal day Depth (m) 50 100 150 200 250 300 350 −2.5 −2 −1.5 −1 −0.5 −20 −15 −10 −5 5 10 15 20 %

Hydraulic ¡redistribu5on ¡affects ¡ seasonal ¡soil ¡moisture ¡(Blodgee) ¡

14 ¡

Weeer ¡ Drier ¡

CLM4.5HD ¡– ¡CLM4.5 ¡

slide-15
SLIDE 15

5 10 15 20 25 30 35 Blodgett Soil Moisture @ 10cm CLM4.5

  • Obs

Month Soil Moisture (%) 1 2 3 4 5 6 7 8 9 10 11 12

slide-16
SLIDE 16

Tapajos ¡site ¡

From: ¡hep://hydrodictyon.eeb.uconn.edu/ ¡ ¡ From: ¡hep://daac.ornl.gov/LBA ¡

slide-17
SLIDE 17

2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4 4.2 CLM4.5

  • 1

2 3 4 5 6 7 8 9 10 11 12 Month ET (mm/day)

Tapajos ¡Site ¡LH ¡Evalua5on ¡

slide-18
SLIDE 18

Impacts ¡on ¡Soil ¡Moisture ¡

20 25 30 35 40 45 Tapajos KM83 Soil Moisture @ 10cm CLM4.5

  • Month

Soil Moisture (%)

  • 1

2 3 4 5 6 7 8 9 10 11 12

slide-19
SLIDE 19

10 20

  • Prec. (mm day−1)

Ordinal day Depth (m) 50 100 150 200 250 300 350 −2.5 −2 −1.5 −1 −0.5 −4 −3 −2 −1 1 2 3 4 %

Hydraulic ¡Redistribu5on ¡Affects ¡ Seasonal ¡Soil ¡Moisture ¡(Tapajos) ¡

CLM4.5HD ¡– ¡CLM4.5 ¡

Weeer ¡ Drier ¡

slide-20
SLIDE 20

CLM ¡Amazon ¡Hydrology ¡ ¡

  • Pedotransfer ¡func5on ¡based ¡on ¡Cosby ¡et ¡al. ¡or ¡

Noihan ¡and ¡Lacarrere ¡(1995) ¡do ¡poorly ¡for ¡ Amazon ¡soils ¡(Delire ¡et ¡al. ¡1997) ¡

  • CLM ¡underes5mates ¡clay ¡frac5on ¡in ¡Amazon ¡
  • No ¡account ¡in ¡CLM ¡for ¡preferen5al ¡flow, ¡

which ¡can ¡be ¡important ¡

  • Differences ¡in ¡climate ¡forcing ¡

8/13/14 ¡ 20 ¡

slide-21
SLIDE 21

Impacts ¡of ¡Oxisols ¡in ¡the ¡Tropics ¡

20oS 10oS 0o 10oN 20oN (b) JJA ∆ET (mm day−1) −0.1 −0.05 0.05 0.1 0.15 180oW 120oW 60oW 0o 60oE 120oE 180oW

slide-22
SLIDE 22

Impacts ¡of ¡RHD ¡in ¡CLM4.5 ¡Compared ¡to ¡ FLUXNET-­‑MTE ¡LH ¡Flux ¡

60oS 30oS 0o 30oN 60oN (b1) 18oS 12oS 6oS 0o 6oN 12oN (b2) 60oS 30oS 0o 30oN 60oN (a1) 18oS 12oS 6oS 0o 6oN 12oN (a2) 180oW 120oW 60oW 0o 60oE 120oE 180oW 60oS 30oS 0o 30oN 60oN (c1) 84oW 72oW 60oW 48oW 36oW 18oS 12oS 6oS 0o 6oN 12oN (c2) −0.8 −0.6 −0.4 −0.2 0.2 0.4 0.6 0.8

CLM4.5 ¡NCEP ¡Forcing ¡ CLM4.5 ¡NCEPForcing+RHD ¡ CLM4.5 ¡CRUNCEP ¡Forcing+RHD ¡

slide-23
SLIDE 23

Summary ¡

  • Blodgee ¡Forest ¡

– Correct ¡numerical ¡solu5on ¡gave ¡poorer ¡fit ¡to ¡LH ¡observa5ons ¡ – RHD ¡improved ¡fit ¡to ¡LH ¡but ¡gave ¡poorer ¡soil ¡moisture ¡ predic5on ¡

  • Comparable ¡to ¡Amenu ¡and ¡Kumar ¡(2008) ¡results ¡
  • Tapajos ¡and ¡Tropics ¡

– Deep ¡roots ¡improved ¡ET ¡seasonal ¡cycle ¡ – Oxisol ¡pedotransfer ¡func5on ¡resulted ¡in ¡~small ¡ET ¡change ¡

  • Climate ¡forcing ¡has ¡large ¡impact ¡on ¡interpreta5on ¡of ¡

mechanisms ¡

  • For ¡a ¡full ¡hydrological ¡evalua5on, ¡CLM ¡needs ¡

restructuring ¡to ¡account ¡for ¡flexible ¡formula5ons ¡of ¡ pedotransfer ¡func5on, ¡root ¡depth ¡profile, ¡soil ¡resistance, ¡ root ¡water ¡uptake, ¡etc. ¡

23 ¡

slide-24
SLIDE 24

Acknowledgements ¡

  • This ¡work ¡was ¡supported ¡by ¡the ¡U.S. ¡DOE ¡

8/13/14 ¡ 24 ¡

slide-25
SLIDE 25

LH ¡Evalua5on ¡Against ¡FLUXNET-­‑MTE ¡

25 ¡

compared ¡to ¡FLUXNET-­‑MTE ¡ Jinyun: ¡ ¡

  • 1. what ¡component ¡of ¡

the ¡forcing ¡creates ¡ the ¡~10 ¡W/m2 ¡ difference ¡in ¡LH ¡ between ¡Qian ¡and ¡ CRU ¡forcing ¡around ¡0 ¡ degrees? ¡

  • 2. HD ¡has ¡higher ¡bias ¡

than ¡default. ¡Why? ¡ How ¡do ¡we ¡argue ¡ about ¡the ¡benefits ¡of ¡ including ¡HD ¡if ¡it ¡ makes ¡the ¡simula5on ¡ worse? ¡

  • 3. Lee ¡et ¡al ¡get ¡~40% ¡

increase ¡in ¡dry ¡season ¡

  • LH. ¡Is ¡that ¡about ¡the ¡

same ¡here? ¡

  • 4. Lee ¡et ¡al ¡also ¡argue ¡

that ¡greater ¡storage ¡

  • f ¡soil ¡water ¡at ¡depth ¡