Benefits and pitfalls of GRACE data assimilation: A case study of terrestrial water storage depletion in India

Manuela Girotto; Gabriëlle J.M. De Lannoy; Rolf H. Reichle; Matthew Rodell; Clara Draper; Soumendra N. Bhanja; Abhijit Mukherjee

doi:10.1002/2017GL072994

Benefits and pitfalls of GRACE data assimilation: A case study of terrestrial water storage depletion in India

Manuela Girotto
, Gabriëlle J.M. De Lannoy
, Rolf H. Reichle
, Matthew Rodell
, Clara Draper
, Soumendra N. Bhanja
, Abhijit Mukherjee

Research output: Contribution to journal › Article › peer-review

117 Scopus citations

Abstract

This study investigates some of the benefits and drawbacks of assimilating terrestrial water storage (TWS) observations from the Gravity Recovery and Climate Experiment (GRACE) into a land surface model over India. GRACE observes TWS depletion associated with anthropogenic groundwater extraction in northwest India. The model, however, does not represent anthropogenic groundwater withdrawals and is not skillful in reproducing the interannual variability of groundwater. Assimilation of GRACE TWS introduces long-term trends and improves the interannual variability in groundwater. But the assimilation also introduces a negative trend in simulated evapotranspiration, whereas in reality evapotranspiration is likely enhanced by irrigation, which is also unmodeled. Moreover, in situ measurements of shallow groundwater show no trend, suggesting that the trends are erroneously introduced by the assimilation into the modeled shallow groundwater, when in reality the groundwater is depleted in deeper aquifers. The results emphasize the importance of representing anthropogenic processes in land surface modeling and data assimilation systems.

Original language	English
Pages (from-to)	4107-4115
Number of pages	9
Journal	Geophysical Research Letters
Volume	44
Issue number	9
DOIs	https://doi.org/10.1002/2017GL072994
State	Published - May 16 2017
Externally published	Yes

Funding

This study was supported by the NASA Terrestrial Hydrology and GRACE and GRACE Follow-On Science Team programs. The GRACE observations used here are from http://grace.jpl.nasa.gov, which is supported by the NASA MEaSUREs Program. This manuscript uses open-source data of the Central Groundwater Board (CGWB), Ministry of Water Resources, River Development and Ganga Rejuvenation, Government of India. S. Bhanja was supported by the SPM fellowship, CSIR (India). A. Mukherjee was supported by the Ministry of Human Resource Development (MHRD) project AGI (IIT/SRIC/GG and CSE/AGI/2013-14/201), application of artificial intelligence in groundwater storage estimation of Indian subcontinent. Computational resources were provided by the NASA High-End Computing Program through the NASA Center for Climate Simulation at the Goddard Space Flight Center.

Keywords

GRACE TWS
India
anthropogenic effects
data assimilation
groundwater
unmodeled processes

Access to Document

10.1002/2017GL072994

Cite this

@article{65f491f2c5d04bdba2da7d4d2e7555d5,

title = "Benefits and pitfalls of GRACE data assimilation: A case study of terrestrial water storage depletion in India",

abstract = "This study investigates some of the benefits and drawbacks of assimilating terrestrial water storage (TWS) observations from the Gravity Recovery and Climate Experiment (GRACE) into a land surface model over India. GRACE observes TWS depletion associated with anthropogenic groundwater extraction in northwest India. The model, however, does not represent anthropogenic groundwater withdrawals and is not skillful in reproducing the interannual variability of groundwater. Assimilation of GRACE TWS introduces long-term trends and improves the interannual variability in groundwater. But the assimilation also introduces a negative trend in simulated evapotranspiration, whereas in reality evapotranspiration is likely enhanced by irrigation, which is also unmodeled. Moreover, in situ measurements of shallow groundwater show no trend, suggesting that the trends are erroneously introduced by the assimilation into the modeled shallow groundwater, when in reality the groundwater is depleted in deeper aquifers. The results emphasize the importance of representing anthropogenic processes in land surface modeling and data assimilation systems.",

keywords = "GRACE TWS, India, anthropogenic effects, data assimilation, groundwater, unmodeled processes",

author = "Manuela Girotto and De Lannoy, \{Gabri{\"e}lle J.M.\} and Reichle, \{Rolf H.\} and Matthew Rodell and Clara Draper and Bhanja, \{Soumendra N.\} and Abhijit Mukherjee",

year = "2017",

month = may,

day = "16",

doi = "10.1002/2017GL072994",

language = "English",

volume = "44",

pages = "4107--4115",

journal = "Geophysical Research Letters",

issn = "0094-8276",

publisher = "Wiley Open Access",

number = "9",

}

TY - JOUR

T1 - Benefits and pitfalls of GRACE data assimilation

T2 - A case study of terrestrial water storage depletion in India

AU - Girotto, Manuela

AU - De Lannoy, Gabriëlle J.M.

AU - Reichle, Rolf H.

AU - Rodell, Matthew

AU - Draper, Clara

AU - Bhanja, Soumendra N.

AU - Mukherjee, Abhijit

PY - 2017/5/16

Y1 - 2017/5/16

N2 - This study investigates some of the benefits and drawbacks of assimilating terrestrial water storage (TWS) observations from the Gravity Recovery and Climate Experiment (GRACE) into a land surface model over India. GRACE observes TWS depletion associated with anthropogenic groundwater extraction in northwest India. The model, however, does not represent anthropogenic groundwater withdrawals and is not skillful in reproducing the interannual variability of groundwater. Assimilation of GRACE TWS introduces long-term trends and improves the interannual variability in groundwater. But the assimilation also introduces a negative trend in simulated evapotranspiration, whereas in reality evapotranspiration is likely enhanced by irrigation, which is also unmodeled. Moreover, in situ measurements of shallow groundwater show no trend, suggesting that the trends are erroneously introduced by the assimilation into the modeled shallow groundwater, when in reality the groundwater is depleted in deeper aquifers. The results emphasize the importance of representing anthropogenic processes in land surface modeling and data assimilation systems.

AB - This study investigates some of the benefits and drawbacks of assimilating terrestrial water storage (TWS) observations from the Gravity Recovery and Climate Experiment (GRACE) into a land surface model over India. GRACE observes TWS depletion associated with anthropogenic groundwater extraction in northwest India. The model, however, does not represent anthropogenic groundwater withdrawals and is not skillful in reproducing the interannual variability of groundwater. Assimilation of GRACE TWS introduces long-term trends and improves the interannual variability in groundwater. But the assimilation also introduces a negative trend in simulated evapotranspiration, whereas in reality evapotranspiration is likely enhanced by irrigation, which is also unmodeled. Moreover, in situ measurements of shallow groundwater show no trend, suggesting that the trends are erroneously introduced by the assimilation into the modeled shallow groundwater, when in reality the groundwater is depleted in deeper aquifers. The results emphasize the importance of representing anthropogenic processes in land surface modeling and data assimilation systems.

KW - GRACE TWS

KW - India

KW - anthropogenic effects

KW - data assimilation

KW - groundwater

KW - unmodeled processes

UR - https://www.scopus.com/pages/publications/85018423720

U2 - 10.1002/2017GL072994

DO - 10.1002/2017GL072994

M3 - Article

AN - SCOPUS:85018423720

SN - 0094-8276

VL - 44

SP - 4107

EP - 4115

JO - Geophysical Research Letters

JF - Geophysical Research Letters

IS - 9

ER -

Benefits and pitfalls of GRACE data assimilation: A case study of terrestrial water storage depletion in India

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this