Application of Bayesian statistics to estimate nitrous oxide emission factors of three nitrogen fertilisers on UK grasslands

N. Cowan, P. Levy, J. Drewer, A. Carswell, R. Shaw, I. Simmons, C. Bache, J. Marinheiro, J. Brichet, A. R. Sanchez-Rodriguez, J. Cotton, P. W. Hill, D. R. Chadwick, D. L. Jones, T. H. Misselbrook, U. Skiba

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Trapezoidal integration by linear interpolation of data points is by far the most commonly used method of cumulative flux calculations of nitrous oxide (N2O)in studies that use flux chambers; however, this method is incapable of providing accurate uncertainty estimates. A Bayesian approach was used to calculate N2O emission factors (EFs)and their associated uncertainties from flux chamber measurements made after the application of nitrogen fertilisers, in the form of ammonium nitrate (AN), urea (Ur)and urea treated with Agrotain® urease inhibitor (UI)at four grassland sites in the UK. The comparison between the cumulative fluxes estimated using the Bayesian and linear interpolation methods were broadly similar (R2 = 0.79); however, the Bayesian method was capable of providing realistic uncertainties when a limited number of data points is available. The study reports mean EF values (and 95% confidence intervals)of 0.60 ± 0.63, 0.29 ± 0.22 and 0.26 ± 0.17% of applied N emitted as N2O for the AN, Ur and UI treatments, respectively. There was no significant difference between N2O emissions from the Ur and UI treatments. In the case of the automatic chamber data collected at one site in this study, the data did not fit the log-normal model, implying that more complex models may be needed, particularly for measurement data with high temporal resolution.

Original languageEnglish
Pages (from-to)362-370
Number of pages9
JournalEnvironment International
Volume128
DOIs
Publication statusPublished - Jul 2019

Fingerprint

nitrous oxide
grassland
urea
fertilizer
nitrogen
flux chamber
inhibitor
ammonium nitrate
interpolation
confidence interval
emission factor
statistics
method

Cite this

Cowan, N. ; Levy, P. ; Drewer, J. ; Carswell, A. ; Shaw, R. ; Simmons, I. ; Bache, C. ; Marinheiro, J. ; Brichet, J. ; Sanchez-Rodriguez, A. R. ; Cotton, J. ; Hill, P. W. ; Chadwick, D. R. ; Jones, D. L. ; Misselbrook, T. H. ; Skiba, U. / Application of Bayesian statistics to estimate nitrous oxide emission factors of three nitrogen fertilisers on UK grasslands. In: Environment International. 2019 ; Vol. 128. pp. 362-370.
@article{2bba6ebd8f1b471da024535b01b578bc,
title = "Application of Bayesian statistics to estimate nitrous oxide emission factors of three nitrogen fertilisers on UK grasslands",
abstract = "Trapezoidal integration by linear interpolation of data points is by far the most commonly used method of cumulative flux calculations of nitrous oxide (N2O)in studies that use flux chambers; however, this method is incapable of providing accurate uncertainty estimates. A Bayesian approach was used to calculate N2O emission factors (EFs)and their associated uncertainties from flux chamber measurements made after the application of nitrogen fertilisers, in the form of ammonium nitrate (AN), urea (Ur)and urea treated with Agrotain{\circledR} urease inhibitor (UI)at four grassland sites in the UK. The comparison between the cumulative fluxes estimated using the Bayesian and linear interpolation methods were broadly similar (R2 = 0.79); however, the Bayesian method was capable of providing realistic uncertainties when a limited number of data points is available. The study reports mean EF values (and 95{\%} confidence intervals)of 0.60 ± 0.63, 0.29 ± 0.22 and 0.26 ± 0.17{\%} of applied N emitted as N2O for the AN, Ur and UI treatments, respectively. There was no significant difference between N2O emissions from the Ur and UI treatments. In the case of the automatic chamber data collected at one site in this study, the data did not fit the log-normal model, implying that more complex models may be needed, particularly for measurement data with high temporal resolution.",
keywords = "Agriculture, NO, Uncertainty, Urea, Urease inhibitor",
author = "N. Cowan and P. Levy and J. Drewer and A. Carswell and R. Shaw and I. Simmons and C. Bache and J. Marinheiro and J. Brichet and Sanchez-Rodriguez, {A. R.} and J. Cotton and Hill, {P. W.} and Chadwick, {D. R.} and Jones, {D. L.} and Misselbrook, {T. H.} and U. Skiba",
year = "2019",
month = "7",
doi = "10.1016/j.envint.2019.04.054",
language = "English",
volume = "128",
pages = "362--370",
journal = "Environment International",
issn = "0160-4120",
publisher = "Pergamon",

}

Cowan, N, Levy, P, Drewer, J, Carswell, A, Shaw, R, Simmons, I, Bache, C, Marinheiro, J, Brichet, J, Sanchez-Rodriguez, AR, Cotton, J, Hill, PW, Chadwick, DR, Jones, DL, Misselbrook, TH & Skiba, U 2019, 'Application of Bayesian statistics to estimate nitrous oxide emission factors of three nitrogen fertilisers on UK grasslands' Environment International, vol. 128, pp. 362-370. https://doi.org/10.1016/j.envint.2019.04.054

Application of Bayesian statistics to estimate nitrous oxide emission factors of three nitrogen fertilisers on UK grasslands. / Cowan, N.; Levy, P.; Drewer, J.; Carswell, A.; Shaw, R.; Simmons, I.; Bache, C.; Marinheiro, J.; Brichet, J.; Sanchez-Rodriguez, A. R.; Cotton, J.; Hill, P. W.; Chadwick, D. R.; Jones, D. L.; Misselbrook, T. H.; Skiba, U.

In: Environment International, Vol. 128, 07.2019, p. 362-370.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Application of Bayesian statistics to estimate nitrous oxide emission factors of three nitrogen fertilisers on UK grasslands

AU - Cowan, N.

AU - Levy, P.

AU - Drewer, J.

AU - Carswell, A.

AU - Shaw, R.

AU - Simmons, I.

AU - Bache, C.

AU - Marinheiro, J.

AU - Brichet, J.

AU - Sanchez-Rodriguez, A. R.

AU - Cotton, J.

AU - Hill, P. W.

AU - Chadwick, D. R.

AU - Jones, D. L.

AU - Misselbrook, T. H.

AU - Skiba, U.

PY - 2019/7

Y1 - 2019/7

N2 - Trapezoidal integration by linear interpolation of data points is by far the most commonly used method of cumulative flux calculations of nitrous oxide (N2O)in studies that use flux chambers; however, this method is incapable of providing accurate uncertainty estimates. A Bayesian approach was used to calculate N2O emission factors (EFs)and their associated uncertainties from flux chamber measurements made after the application of nitrogen fertilisers, in the form of ammonium nitrate (AN), urea (Ur)and urea treated with Agrotain® urease inhibitor (UI)at four grassland sites in the UK. The comparison between the cumulative fluxes estimated using the Bayesian and linear interpolation methods were broadly similar (R2 = 0.79); however, the Bayesian method was capable of providing realistic uncertainties when a limited number of data points is available. The study reports mean EF values (and 95% confidence intervals)of 0.60 ± 0.63, 0.29 ± 0.22 and 0.26 ± 0.17% of applied N emitted as N2O for the AN, Ur and UI treatments, respectively. There was no significant difference between N2O emissions from the Ur and UI treatments. In the case of the automatic chamber data collected at one site in this study, the data did not fit the log-normal model, implying that more complex models may be needed, particularly for measurement data with high temporal resolution.

AB - Trapezoidal integration by linear interpolation of data points is by far the most commonly used method of cumulative flux calculations of nitrous oxide (N2O)in studies that use flux chambers; however, this method is incapable of providing accurate uncertainty estimates. A Bayesian approach was used to calculate N2O emission factors (EFs)and their associated uncertainties from flux chamber measurements made after the application of nitrogen fertilisers, in the form of ammonium nitrate (AN), urea (Ur)and urea treated with Agrotain® urease inhibitor (UI)at four grassland sites in the UK. The comparison between the cumulative fluxes estimated using the Bayesian and linear interpolation methods were broadly similar (R2 = 0.79); however, the Bayesian method was capable of providing realistic uncertainties when a limited number of data points is available. The study reports mean EF values (and 95% confidence intervals)of 0.60 ± 0.63, 0.29 ± 0.22 and 0.26 ± 0.17% of applied N emitted as N2O for the AN, Ur and UI treatments, respectively. There was no significant difference between N2O emissions from the Ur and UI treatments. In the case of the automatic chamber data collected at one site in this study, the data did not fit the log-normal model, implying that more complex models may be needed, particularly for measurement data with high temporal resolution.

KW - Agriculture

KW - NO

KW - Uncertainty

KW - Urea

KW - Urease inhibitor

UR - http://www.scopus.com/inward/record.url?scp=85065165471&partnerID=8YFLogxK

U2 - 10.1016/j.envint.2019.04.054

DO - 10.1016/j.envint.2019.04.054

M3 - Article

VL - 128

SP - 362

EP - 370

JO - Environment International

JF - Environment International

SN - 0160-4120

ER -