Empirical and dynamic approaches for modelling the yield and N content of European grasslands

Martha Dellar, Cairistiona Topp, Guillermo Pardo, Agustin del Prado, Nuala Fitton, David Holmes, Gerogios Banos, Eileen Wall

Research output: Contribution to journalArticle

Abstract

We applied two approaches to model grassland yield and nitrogen (N) content. The first was a series of regression equations; the second was the Century dynamic model. The regression model was generated from data from eighty-nine experimental sites across Europe, distinguishing between five climatic regions. The Century model was applied to six sites across these regions. Both approaches estimated mean grassland yields and N content reasonably well, though the root mean squared error tended to be lower for the dynamic model. The regression model achieved better correlations between observed and predicted values. Both models were more sensitive to uncertainties in weather than in soil properties, with precipitation often accounting for the majority of model uncertainty. The regression approach is applicable over large spatial scales but lacks precision, making it suitable for considering general trends. Century is better applied at a local level where more detailed and specific analysis is required.

Original languageEnglish
Article number104562
JournalEnvironmental Modelling and Software
Volume122
Early online date14 Oct 2019
DOIs
Publication statusPublished - Dec 2019

Fingerprint

grassland
modeling
Dynamic models
climatic region
Nitrogen
Soils
soil property
weather
nitrogen
Uncertainty

Keywords

  • grasslands
  • yield
  • nitrogen
  • modelling
  • Grasslands
  • CALIBRATION
  • PERFORMANCE
  • SENSITIVITY
  • Nitrogen
  • MEADOW
  • BIOMASS
  • FERTILIZER
  • UNCERTAINTY
  • NITROGEN
  • SOIL
  • Yield
  • Modelling
  • IMPACTS

Cite this

Empirical and dynamic approaches for modelling the yield and N content of European grasslands. / Dellar, Martha; Topp, Cairistiona; Pardo, Guillermo; del Prado, Agustin; Fitton, Nuala; Holmes, David; Banos, Gerogios; Wall, Eileen.

In: Environmental Modelling and Software, Vol. 122, 104562, 12.2019.

Research output: Contribution to journalArticle

Dellar, Martha ; Topp, Cairistiona ; Pardo, Guillermo ; del Prado, Agustin ; Fitton, Nuala ; Holmes, David ; Banos, Gerogios ; Wall, Eileen. / Empirical and dynamic approaches for modelling the yield and N content of European grasslands. In: Environmental Modelling and Software. 2019 ; Vol. 122.
@article{dec5f3f32017424eb466e364db856afa,
title = "Empirical and dynamic approaches for modelling the yield and N content of European grasslands",
abstract = "We applied two approaches to model grassland yield and nitrogen (N) content. The first was a series of regression equations; the second was the Century dynamic model. The regression model was generated from data from eighty-nine experimental sites across Europe, distinguishing between five climatic regions. The Century model was applied to six sites across these regions. Both approaches estimated mean grassland yields and N content reasonably well, though the root mean squared error tended to be lower for the dynamic model. The regression model achieved better correlations between observed and predicted values. Both models were more sensitive to uncertainties in weather than in soil properties, with precipitation often accounting for the majority of model uncertainty. The regression approach is applicable over large spatial scales but lacks precision, making it suitable for considering general trends. Century is better applied at a local level where more detailed and specific analysis is required.",
keywords = "grasslands, yield, nitrogen, modelling, Grasslands, CALIBRATION, PERFORMANCE, SENSITIVITY, Nitrogen, MEADOW, BIOMASS, FERTILIZER, UNCERTAINTY, NITROGEN, SOIL, Yield, Modelling, IMPACTS",
author = "Martha Dellar and Cairistiona Topp and Guillermo Pardo and {del Prado}, Agustin and Nuala Fitton and David Holmes and Gerogios Banos and Eileen Wall",
note = "This work was supported by the Horizon 2020 SFS-01c-2015 project entitled “Innovation of sustainable sheep and goat production in Europe (iSAGE)” [grant number 679302]; and the Rural & Environment Science & Analytical Services Division of the Scottish Government. BC3 is supported by the Basque Government through the BERC 2018–2021 program and by Spanish Ministry of Economy and Competitiveness MINECO through BC3 Mar{\'i}a de Maeztu excellence accreditation MDM-2017-0714. Agustin del Prado is supported by the Ramon y Cajal Programme. We would like to thank all the people who provided the data which made this work possible. In particular, Professor Wolfgang Schmidt, for data from the Experimental Botanical Garden of G{\"o}ttingen University. Also the Lawes Agricultural Trust and Rothamsted Research for data from the e-RA database. The Rothamsted Long-term Experiments National Capability (LTE-NCG) is supported by the UK Biotechnology and Biological Sciences Research Council and the Lawes Agricultural Trust.",
year = "2019",
month = "12",
doi = "10.1016/j.envsoft.2019.104562",
language = "English",
volume = "122",
journal = "Environmental Modelling and Software",
issn = "1364-8152",
publisher = "ELSEVIER APPL SCI PUBL LTD",

}

TY - JOUR

T1 - Empirical and dynamic approaches for modelling the yield and N content of European grasslands

AU - Dellar, Martha

AU - Topp, Cairistiona

AU - Pardo, Guillermo

AU - del Prado, Agustin

AU - Fitton, Nuala

AU - Holmes, David

AU - Banos, Gerogios

AU - Wall, Eileen

N1 - This work was supported by the Horizon 2020 SFS-01c-2015 project entitled “Innovation of sustainable sheep and goat production in Europe (iSAGE)” [grant number 679302]; and the Rural & Environment Science & Analytical Services Division of the Scottish Government. BC3 is supported by the Basque Government through the BERC 2018–2021 program and by Spanish Ministry of Economy and Competitiveness MINECO through BC3 María de Maeztu excellence accreditation MDM-2017-0714. Agustin del Prado is supported by the Ramon y Cajal Programme. We would like to thank all the people who provided the data which made this work possible. In particular, Professor Wolfgang Schmidt, for data from the Experimental Botanical Garden of Göttingen University. Also the Lawes Agricultural Trust and Rothamsted Research for data from the e-RA database. The Rothamsted Long-term Experiments National Capability (LTE-NCG) is supported by the UK Biotechnology and Biological Sciences Research Council and the Lawes Agricultural Trust.

PY - 2019/12

Y1 - 2019/12

N2 - We applied two approaches to model grassland yield and nitrogen (N) content. The first was a series of regression equations; the second was the Century dynamic model. The regression model was generated from data from eighty-nine experimental sites across Europe, distinguishing between five climatic regions. The Century model was applied to six sites across these regions. Both approaches estimated mean grassland yields and N content reasonably well, though the root mean squared error tended to be lower for the dynamic model. The regression model achieved better correlations between observed and predicted values. Both models were more sensitive to uncertainties in weather than in soil properties, with precipitation often accounting for the majority of model uncertainty. The regression approach is applicable over large spatial scales but lacks precision, making it suitable for considering general trends. Century is better applied at a local level where more detailed and specific analysis is required.

AB - We applied two approaches to model grassland yield and nitrogen (N) content. The first was a series of regression equations; the second was the Century dynamic model. The regression model was generated from data from eighty-nine experimental sites across Europe, distinguishing between five climatic regions. The Century model was applied to six sites across these regions. Both approaches estimated mean grassland yields and N content reasonably well, though the root mean squared error tended to be lower for the dynamic model. The regression model achieved better correlations between observed and predicted values. Both models were more sensitive to uncertainties in weather than in soil properties, with precipitation often accounting for the majority of model uncertainty. The regression approach is applicable over large spatial scales but lacks precision, making it suitable for considering general trends. Century is better applied at a local level where more detailed and specific analysis is required.

KW - grasslands

KW - yield

KW - nitrogen

KW - modelling

KW - Grasslands

KW - CALIBRATION

KW - PERFORMANCE

KW - SENSITIVITY

KW - Nitrogen

KW - MEADOW

KW - BIOMASS

KW - FERTILIZER

KW - UNCERTAINTY

KW - NITROGEN

KW - SOIL

KW - Yield

KW - Modelling

KW - IMPACTS

U2 - 10.1016/j.envsoft.2019.104562

DO - 10.1016/j.envsoft.2019.104562

M3 - Article

VL - 122

JO - Environmental Modelling and Software

JF - Environmental Modelling and Software

SN - 1364-8152

M1 - 104562

ER -