Evaluating lysimeter drainage against soil deep percolation modeled with profile soil moisture and lab measured hydraulic properties

Hydrological Observatory

www.hobecenter.dk

Vicente Vásquez, Kirsten Schelde, Anton Thomsen and Bo V. Iversen Dept. of Agroecology, Aarhus University, DK-8830 Tjele, Denmark

Introduction Quantifying recharge via deep percolation is needed for sustainable

management of water resources The impact of climate change on recharge can be addressed with robust

modeled predictions Combining different techniques for in situ determination of soil deep

percolation can reduce the uncertainties on water balance

Research questions Can soil water flow modeling based on lab measured hydraulic properties

replicate field measurements of profile soil moisture? Is there enough information in soil texture and profile soil water content

time series to estimate soil unsaturated flow? How big is the difference in estimated annual recharge between

lysimeters, inverse modeling of profile soil moisture and texture (SWC METHOD), and forward modeling based on lab measured hydraulic properties (LAB METHOD)?

The 1D soil water flow model Hydrus 1D based on Richards equation and the Mualem (1976) van

Genuchten (1980) model Inverse optimization of soil hydraulic parameters α, Ksat with profile soil

moisture time series, texture and atmospheric (transient) boundary conditions. All other parameters kept constant to the default textural values available in Hydrus 1D.

Measurements Lab determination of water retention curve, Ksat and K(h) with drip

infiltrometer on large intact soil cores (20x20cm) Automated large (12.5 m2) zero tension lysimeters located below the root

zone. Automated profile TDR measurements for below root zone observations of

soil moisture; capacitance probes for higher depth resolution near the surface

Materials and methods

Field instrumentation

Four 12.5m2 repacked lysimeters Study area: A winter barley field in Skjern river catchment Yellow rectangle is the area for instrumentation of agricultural field observatory. Blue rectangle shows the location of four lysimeters. Red rectangle shows location of Profile TDR

Upper boundary and root zone

Figure 2. Cumulative ETec against residual ET from water balance in lysimeters

Highlights Near surface soil moisture is overestimated (0-30cm) and

underestimated (30-60cm) by LAB METHOD, but the overall dynamics are well represented

Annual recharge estimated at 60 cm and 2m depth with SWC METHOD was 17 and 20% (100-120 mm) higher than LAB METHOD, probably due to uncertainty in estimation of Ksat from texture.

Annual recharge from lysimeters and SWC METHOD (2m depth) is in good agreement (< 5mm), however; lysimeters yield higher flow rates during high precipitation

In general, recharge from LAB METHOD was 17% lower than lysimeters and SWC METHOD. This is probably an indication that even in sand soils (>90% sand), differences between repacked soils (SWC METHOD and Lysimeters) and undisturbed soils (intact soil cores-LAB METHOD) have a large impact on the estimated recharge

Good agreement between lysimeters and SWC METHOD exist, validating the use of this method for upscaling ground water recharge

Below root zone field, lysimeters

Conclusions

Soil moisture time series and soil texture information can be used effectively to predict ground water recharge

In situ soil moisture dynamics was well represented by

lab hydraulic property cores, however, the magnitude of soil moisture measurements was not

Annual recharge with a combination of independent

methods had an uncertainty of ±10% References

Mualem Y. (1976) New Model for Predicting Hydraulic Conductivity of Unsaturated Porous-Media. Water Resources Research 12:513-522. Simunek J., van Genuchten M.T., Sejna M. (2008) Development and applications of the HYDRUS and STANMOD software packages and related codes. Vadose Zone Journal 7:587-600. van Genuchten, M. Th. 1980. A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. SoilSci. Soc. Am. J. 44:892-898.

Lab measured hydraulic properties undisturbed soil cores

Cumulative Drainage inferred from 2m SWC + texture, mm

0 200 400 600 800C

umul

ativ

e D

rain

age

Lysi

met

ers

and

Mod

eled

Lab

cor

es, m

m

0

200

400

600

800

LAB METHOD at 2m Lysimeters at 2m

2011-2012

Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep

Lysi

met

er d

rain

age,

mm

/day

0

10

20

30

40

50

60

Cum

ulat

ive

Lysi

met

er d

rain

age,

mm

0

200

400

600

800

Lysimeters at 2m

2011-2012

Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep

Dra

inag

e, m

m/d

ay

0

10

20

30

40

50

60

Cum

ulat

ive

drai

nage

, mm

0

200

400

600

800SWC METHOD at 2m LAB METHOD at 2m

2011-2012

Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep

Dra

inag

e, m

m/d

ay

0

10

20

30

40

50

60

Cum

ulat

ive

drai

nage

D, m

m

-200

0

200

400

600

800SWC METHOD at 60 LAB METHOD at 60 cm

2011-2012

Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep

Vol

umet

ric s

oil w

ater

con

tent

SW

C, c

m3/

cm3

0.0

0.1

0.2

0.3

0.4

0.5

0-30 cm measured30-60 cm measured0-30cm LAB METHOD30-60cm LAB METHOD

2011-2012

Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep

Ref

eren

ce E

vapo

trans

pira

tion,

mm

0

2

4

6

8

10

12

Pre

cipi

tatio

n, m

m

0

20

40

60

80

LAI

0

2

4

6

8

ETrefprecipLAI

UID: 76790