See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/309390624

Determination of clogging during managed aquifer recharge operation by

laboratory experiments and mathematical models

Conference Paper · September 2016

CITATION READS

1 110

3 authors:

Jana Glass Thomas Fichtner

TU Dresden TU Dresden
42 PUBLICATIONS 237 CITATIONS 50 PUBLICATIONS 94 CITATIONS

SEE PROFILE SEE PROFILE

Catalin Stefan
TU Dresden
132 PUBLICATIONS 1,406 CITATIONS

SEE PROFILE

All content following this page was uploaded by Jana Glass on 29 October 2018.

The user has requested enhancement of the downloaded file.

 Determination of clogging during managed aquifer recharge operation by
laboratory experiments and mathematical models
J. Ringleb, T. Fichtner, C. Stefan
Technische Universität Dresden, Faculty of Environmental Sciences, Department of Hydro Science www.tu-dresden.de/uw/inowas

BACKGROUND & OBJECTIVES METHODS: Clogging determination RESULTS

1.1
• Laboratory tank experiments designed to predict the efficiency of 1) Tracer experiments Pedretti et al. 2012
1
managed aquifer recharge under different processes and site- • KBr as a tracer to determine average flow velocity (R²=0.88)
0.9
specific conditions (t50) in the centre (Position 2, 0.3 m) Libardi et al.1980
• Determination of clogging (infiltration capacity reduction) often a 2) Libardi et al. (1980) (Reichhardt et al. 1998)
0.8
(R²=0.79)
challenge as different processes interact and influence its extent 0.7
• Calculation of unsaturated hydraulic conductivity
 Comparison of different clogging estimation techniques 0.6 HYDRUS-2D

K/Ko
during drying phase using water content (R²=0.61)
0.5
measurements (Position 2, 0.3 m)
METHODS: Laboratory tank experiments 𝐾(𝜃) = 𝐾𝑜 𝑒 −𝛾(𝜃0 −𝜃) 0.4 Pavelic et al. 2007
(R²=0.82)
• Rectangular shaped stainless steel lysimeter (1.0 m / 1.5 m / 1.0 m) 1 1 𝛾𝐾0 𝑧 𝛾𝐵 0.3

placed in a climate tent with controlled temperature (17°C) and 𝜃0 − 𝜃 = 𝑙𝑛 𝑡 + 𝑙𝑛 = 𝐴𝑧 𝑙𝑛 𝑡 + 𝐵𝑧 𝐾0 = 𝑒 𝑧

𝛾 𝛾 𝑧 𝛾 0.2 KBr tracer experiment
humidity (60%) 0.1
3) HYDRUS-2D (Šimůnek et al. 2016)
• Tensiometers, TDR-probes in two depths (0.3 m and 0.7 m)
• Setup of numerical unsaturated flow model based on 0
• Infiltration of river water in wet/dry cycles (6 h:18 h) into infiltration Richards equation along cross section
0 20 40 60
time [days]
80 100 120

basin with constant pumping rate (0.4 m/d) for 120 days Fig. 2. Comparison of ratio of hydraulic conductivity with time determined by different methods. In
• Transient calibration of ks using inverse solution and
• Concentration of DOC (5-36 mg/L) and suspended solids (2-41 mg/L) three days of measured tensiometer and water
addition coefficients of determination (R²) comparing tracer test results with other estimation
in the river water fluctuated over time techniques are given.
content data (Position 2,3 in 0.3 m and 0.7 m) Symbology
4) Pedretti et al. (2012) • Tracer test shows that saturated hydraulic conductivity θ Water content
decreases by 65% during 114 days K Hydraulic conductivity
• Analytical
Faculty of Environmental Sciences, Department of Hydro Sciences formulae to calculate infiltration capacity Ko Initial hydraulic conductivity
• Based on water content measurements the decrease in γ Slope reciprocal of the Libardis water
changes due to physical and biological clogging
−3(λ𝑝 +λ𝑏 )𝑡
hydraulic conductivity calculated with Libardi et al. 1980 content smoothing equation
𝐼 𝑡 = 𝐼0 ∗ 𝑒 is lower (51%) than tracer test estimations z Depth of water content measurement
∗ I Infiltration capacity
𝜌𝑘 𝑀𝑏 λ𝑠 • Pedretti et al. 2012 and Pavelic et al. 2007 slightly
λ𝑝 = 𝑚𝑝 ϑ𝑎 λ𝑧 λ𝑏 = λp Physical clogging factor
𝜌𝑏 𝜑0 overestimate clogging (73% / 78%) compared to tracer λb Biological clogging factor
λz Filtration coefficient
5) Pavelic et al. (2007) test mp Coefficient of proportionality
ϑa Average particle attachment velocity
• Analytical formulae to calculate mass changes due • HYDRUS-2D cannot reproduce the changes in hydraulic λs Microbial growth parameter
to physical and biological clogging and relate it to conductivity due to clogging using measured data for ρK Soil bulk density
calibration ρb Biomass / biofilm density
conductivity changes φ Porosity
1 𝑐0 𝑐𝑠 Mb* Initial biomass distribution (attached
∆𝑀𝑝 𝑡 = 𝑐𝑖 − 𝑐𝑒 𝑄𝑖 ∆𝑡𝑖 ∆𝑀𝑏 𝑡 = 𝑀𝑎
𝑐0 + (𝑐𝑠 − 𝑐0 )𝑒 −𝑘𝑏 𝑡 CONCLUSION to the soil)
∆Mp Mass change due to particle retention
2 ∆𝑀𝑏 𝑡2 ∆𝑀𝑝 𝑡2 • Except HYDRUS-2D, the tested methods estimate ∆Mb Mass change due to biofilm growth
Fig. 1. Setup of laboratory tank 𝜑𝑡2 = 𝜑𝑡1 − − ci Suspended solid concentration inflow
𝜌𝑏 𝜌𝑝 clogging to decrease the hydraulic conductivity in the ce Suspended solid concentration
experiment: top profile view laboratory tank during the conducted experiments outflow
and low top view including 𝜑𝑡2 3 (1 − 𝜑𝑡1 ) 2 co Initial biofilm concentration
𝑘𝑡2 = 𝑘𝑡1 ( ) ( ) • Pedretti et al. 2012 followed by Pavelic et al. 2007 ρp Particles density
3 measurement devices. 𝜑𝑡1 (1 − 𝜑𝑡2 ) match best with tracer test results kb Biofilm growth rate
cs Plateau concentration (biofilm)
Funded by Literature:
Technische Universität Dresden Reichardt et. al 1998. Critical analysis of the field determination of soil hydraulic conductivity functions using the flux-gradient approach. Soil and Tillage Research 48, 81–89. Qi Rate of injection / flux rate through
Pratzschwitzer Str. 15, 01796 Pirna, Germany Pedretti et. al 2012. Probabilistic analysis of maintenance and operation of artificial recharge ponds. Advances in Water Resources 36, 23–35. column
Pavelic et. al 2007. Water quality effects on clogging rates during reclaimed water ASR in a carbonate aquifer. Journal of Hydrology 334, 1–16. Ma Mass of aquifer material within the
Tel. +49 3501 530046; Email: [email protected] Šimůnek et al. 2016. Recent Developments and Applications of the HYDRUS Computer Software Packages. Vadose Zone Journal 15. zone of interest
View publication stats