Towards a comprehensive approach to parameter estimation in land surface parameterization schemes

Rafael Rosolem, Hoshin Vijai Gupta, W. James Shuttleworth, Luis Gustavo Gonçalves de Gonçalves, Xubin Zeng

Research output: Contribution to journalArticle

31 Citations (Scopus)

Abstract

In climate models, the land-atmosphere interactions are described numerically by land surface parameterization (LSP) schemes. The continuing improvement in realism in these schemes comes at the expense of the need to specify a large number of parameters that are either directly measured or estimated. Also, an emerging problem is whether the relationships used in LSPs are universal and globally applicable. One plausible approach to evaluate this is to first minimize uncertainty in model parameters by calibration. In this paper, we conduct a comprehensive analysis of some model diagnostics using a slightly modified version of the Simple Biosphere 3 model for a variety of biomes located mainly in the Amazon. First, the degree of influence of each individual parameter in simulating surface fluxes is identified. Next, we estimate parameters using a multi-operator genetic algorithm applied in a multi-objective context and evaluate simulations of energy and carbon fluxes against observations. Compared with the default parameter sets, these parameter estimates improve the partitioning of energy fluxes in forest and cropland sites and provide better simulations of daytime increases in assimilation of net carbon during the dry season at forest sites. Finally, a detailed assessment of the parameter estimation problem was performed by accounting for the decomposition of the mean squared error to the total model uncertainty. Analysis of the total prediction uncertainty reveals that the parameter adjustments significantly improve reproduction of the mean and variability of the flux time series at all sites and generally remove seasonality of the errors but do not improve dynamical properties. Our results demonstrate that error decomposition provides a meaningful and intuitive way to understand differences in model performance. To make further advancements in the knowledge of these models, we encourage the LSP community to adopt similar approaches in the future.

Original languageEnglish (US)
Pages (from-to)2075-2097
Number of pages23
JournalHydrological Processes
Volume27
Issue number14
DOIs
StatePublished - Jul 1 2013

Fingerprint

land surface
parameterization
energy flux
decomposition
surface flux
carbon flux
biome
parameter
parameter estimation
genetic algorithm
biosphere
simulation
seasonality
dry season
climate modeling
partitioning
time series
calibration
atmosphere
carbon

Keywords

  • Amazon biomes
  • Land surface modelling
  • Mean squared error decomposition
  • Model diagnostics
  • Parameter estimation
  • Simple biosphere model

ASJC Scopus subject areas

  • Water Science and Technology

Cite this

Towards a comprehensive approach to parameter estimation in land surface parameterization schemes. / Rosolem, Rafael; Gupta, Hoshin Vijai; Shuttleworth, W. James; de Gonçalves, Luis Gustavo Gonçalves; Zeng, Xubin.

In: Hydrological Processes, Vol. 27, No. 14, 01.07.2013, p. 2075-2097.

Research output: Contribution to journalArticle

@article{f563a90c9e244f5facf7621d2a11f0fb,
title = "Towards a comprehensive approach to parameter estimation in land surface parameterization schemes",
abstract = "In climate models, the land-atmosphere interactions are described numerically by land surface parameterization (LSP) schemes. The continuing improvement in realism in these schemes comes at the expense of the need to specify a large number of parameters that are either directly measured or estimated. Also, an emerging problem is whether the relationships used in LSPs are universal and globally applicable. One plausible approach to evaluate this is to first minimize uncertainty in model parameters by calibration. In this paper, we conduct a comprehensive analysis of some model diagnostics using a slightly modified version of the Simple Biosphere 3 model for a variety of biomes located mainly in the Amazon. First, the degree of influence of each individual parameter in simulating surface fluxes is identified. Next, we estimate parameters using a multi-operator genetic algorithm applied in a multi-objective context and evaluate simulations of energy and carbon fluxes against observations. Compared with the default parameter sets, these parameter estimates improve the partitioning of energy fluxes in forest and cropland sites and provide better simulations of daytime increases in assimilation of net carbon during the dry season at forest sites. Finally, a detailed assessment of the parameter estimation problem was performed by accounting for the decomposition of the mean squared error to the total model uncertainty. Analysis of the total prediction uncertainty reveals that the parameter adjustments significantly improve reproduction of the mean and variability of the flux time series at all sites and generally remove seasonality of the errors but do not improve dynamical properties. Our results demonstrate that error decomposition provides a meaningful and intuitive way to understand differences in model performance. To make further advancements in the knowledge of these models, we encourage the LSP community to adopt similar approaches in the future.",
keywords = "Amazon biomes, Land surface modelling, Mean squared error decomposition, Model diagnostics, Parameter estimation, Simple biosphere model",
author = "Rafael Rosolem and Gupta, {Hoshin Vijai} and Shuttleworth, {W. James} and {de Gon{\cc}alves}, {Luis Gustavo Gon{\cc}alves} and Xubin Zeng",
year = "2013",
month = "7",
day = "1",
doi = "10.1002/hyp.9362",
language = "English (US)",
volume = "27",
pages = "2075--2097",
journal = "Hydrological Processes",
issn = "0885-6087",
publisher = "John Wiley and Sons Ltd",
number = "14",

}

TY - JOUR

T1 - Towards a comprehensive approach to parameter estimation in land surface parameterization schemes

AU - Rosolem, Rafael

AU - Gupta, Hoshin Vijai

AU - Shuttleworth, W. James

AU - de Gonçalves, Luis Gustavo Gonçalves

AU - Zeng, Xubin

PY - 2013/7/1

Y1 - 2013/7/1

N2 - In climate models, the land-atmosphere interactions are described numerically by land surface parameterization (LSP) schemes. The continuing improvement in realism in these schemes comes at the expense of the need to specify a large number of parameters that are either directly measured or estimated. Also, an emerging problem is whether the relationships used in LSPs are universal and globally applicable. One plausible approach to evaluate this is to first minimize uncertainty in model parameters by calibration. In this paper, we conduct a comprehensive analysis of some model diagnostics using a slightly modified version of the Simple Biosphere 3 model for a variety of biomes located mainly in the Amazon. First, the degree of influence of each individual parameter in simulating surface fluxes is identified. Next, we estimate parameters using a multi-operator genetic algorithm applied in a multi-objective context and evaluate simulations of energy and carbon fluxes against observations. Compared with the default parameter sets, these parameter estimates improve the partitioning of energy fluxes in forest and cropland sites and provide better simulations of daytime increases in assimilation of net carbon during the dry season at forest sites. Finally, a detailed assessment of the parameter estimation problem was performed by accounting for the decomposition of the mean squared error to the total model uncertainty. Analysis of the total prediction uncertainty reveals that the parameter adjustments significantly improve reproduction of the mean and variability of the flux time series at all sites and generally remove seasonality of the errors but do not improve dynamical properties. Our results demonstrate that error decomposition provides a meaningful and intuitive way to understand differences in model performance. To make further advancements in the knowledge of these models, we encourage the LSP community to adopt similar approaches in the future.

AB - In climate models, the land-atmosphere interactions are described numerically by land surface parameterization (LSP) schemes. The continuing improvement in realism in these schemes comes at the expense of the need to specify a large number of parameters that are either directly measured or estimated. Also, an emerging problem is whether the relationships used in LSPs are universal and globally applicable. One plausible approach to evaluate this is to first minimize uncertainty in model parameters by calibration. In this paper, we conduct a comprehensive analysis of some model diagnostics using a slightly modified version of the Simple Biosphere 3 model for a variety of biomes located mainly in the Amazon. First, the degree of influence of each individual parameter in simulating surface fluxes is identified. Next, we estimate parameters using a multi-operator genetic algorithm applied in a multi-objective context and evaluate simulations of energy and carbon fluxes against observations. Compared with the default parameter sets, these parameter estimates improve the partitioning of energy fluxes in forest and cropland sites and provide better simulations of daytime increases in assimilation of net carbon during the dry season at forest sites. Finally, a detailed assessment of the parameter estimation problem was performed by accounting for the decomposition of the mean squared error to the total model uncertainty. Analysis of the total prediction uncertainty reveals that the parameter adjustments significantly improve reproduction of the mean and variability of the flux time series at all sites and generally remove seasonality of the errors but do not improve dynamical properties. Our results demonstrate that error decomposition provides a meaningful and intuitive way to understand differences in model performance. To make further advancements in the knowledge of these models, we encourage the LSP community to adopt similar approaches in the future.

KW - Amazon biomes

KW - Land surface modelling

KW - Mean squared error decomposition

KW - Model diagnostics

KW - Parameter estimation

KW - Simple biosphere model

UR - http://www.scopus.com/inward/record.url?scp=84879883328&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879883328&partnerID=8YFLogxK

U2 - 10.1002/hyp.9362

DO - 10.1002/hyp.9362

M3 - Article

AN - SCOPUS:84879883328

VL - 27

SP - 2075

EP - 2097

JO - Hydrological Processes

JF - Hydrological Processes

SN - 0885-6087

IS - 14

ER -