What does it take to detect a change in soil carbon stock? a regional comparison of minimum detectable difference and experiment duration in the north central united states

M. Necpálová, R. P. Anex, A. N. KxravchenkoProf, L. J. Abendroth, S. J.Del Grosso, W. A. Dickprof, M. J. HelmersProf, D. Herzmann, J. G. LauerProf, E. D. NafzigerProf, J. E. SawyerProf, P. C. ScharfProf, J. S. Strockprof, M. B. VillamilProf

Research output: Contribution to journalArticle

Abstract

Variability in soil organic carbon (SOC) results from natural and human processes interacting across time and space, and leads to large variation in the minimum difference in SOC that can be detected with a particular experimental design. Here we report a unique comparison of minimum detectable differences (MDDs) in SOC, and the estimated times required to observe those MDDs across the north central United States, calculated for the two most common SOC experiments: (1) a comparison between two treatments, e.g., moldboard plow (MP) and no-tillage (NT), using a randomized complete block design experiment; and (2) a comparison of changes in SOC over time for a particular treatment, e.g., NT, using a randomized complete block design experiment with time as an additional factor. We estimated the duration of the two experiment types required to achieve MDD through simulation of SOC dynamics. Data for the study came from 13 experimental sites located in Iowa, Illinois, Ohio, Michigan,Wisconsin, Missouri, and Minnesota. Soil organic carbon, bulk density, and texture were measured at four soil depths. Minimum detectable differences were calculated with probability ofType I error of 0.05 and probability ofType II error of 0.15. The MDDs in SOC were highly variable across the region and increased with soil depth. At 0 to 10 cm (0 to 3.9 in) soil depth, MDDs with five replications ranged from 1.04 g C kg-1 (0.017 oz C lb-1; 6%) to 7.15 g C kg-1 (0.114 oz C lb-1; 31%) for comparison of two treatments; and from 0.46 g C kg-1 (0.007 oz C lb-1; 3%) to 3.12 g C kg-1 (0.050 oz C lb-1; 13%) for SOC change over time. Large differences were also predicted in the experiment duration required to detect a difference in SOC between MP and NT (from 8 to >100 years with five replications), or a change in SOC over time under NT management (from 11 to 71 years with five replications). At most locations, the time required to detect a change in SOC under NT was shorter than the time required to detect a difference between MP and NT. Minimum detectable difference and experiment duration decreased with the number of replications and were correlated with SOC variability and soil texture of the experimental sites, i.e., they tended to be lower in fine textured soils. Experiment duration was also reduced by increased crop productivity and the amount of residue left on the soil. The relationships and methods described here enable the design of experiments with high power of detecting differences and changes in SOC and enhance our understanding of how management practices influence SOC storage.

Original languageEnglish (US)
Pages (from-to)517-531
Number of pages15
JournalJournal of Soil and Water Conservation
Volume69
Issue number6
DOIs
StatePublished - Nov 1 2014

Fingerprint

Midwestern United States
soil carbon
carbon sinks
soil organic carbon
organic carbon
duration
soil
experiment
zero tillage
no-tillage
moldboard plows
experimental design
soil depth
comparison
fine-textured soils
soil texture
carbon sequestration
space and time

Keywords

  • Experiment duration
  • Minimum detectable difference
  • No-till
  • Soil organic carbon
  • Statistical power
  • Tillage

ASJC Scopus subject areas

  • Agronomy and Crop Science
  • Water Science and Technology
  • Soil Science
  • Nature and Landscape Conservation

Cite this

What does it take to detect a change in soil carbon stock? a regional comparison of minimum detectable difference and experiment duration in the north central united states. / Necpálová, M.; Anex, R. P.; KxravchenkoProf, A. N.; Abendroth, L. J.; Grosso, S. J.Del; Dickprof, W. A.; HelmersProf, M. J.; Herzmann, D.; LauerProf, J. G.; NafzigerProf, E. D.; SawyerProf, J. E.; ScharfProf, P. C.; Strockprof, J. S.; VillamilProf, M. B.

In: Journal of Soil and Water Conservation, Vol. 69, No. 6, 01.11.2014, p. 517-531.

Research output: Contribution to journalArticle

Necpálová, M, Anex, RP, KxravchenkoProf, AN, Abendroth, LJ, Grosso, SJD, Dickprof, WA, HelmersProf, MJ, Herzmann, D, LauerProf, JG, NafzigerProf, ED, SawyerProf, JE, ScharfProf, PC, Strockprof, JS & VillamilProf, MB 2014, 'What does it take to detect a change in soil carbon stock? a regional comparison of minimum detectable difference and experiment duration in the north central united states', Journal of Soil and Water Conservation, vol. 69, no. 6, pp. 517-531. https://doi.org/10.2489/jswc.69.6.517
Necpálová, M. ; Anex, R. P. ; KxravchenkoProf, A. N. ; Abendroth, L. J. ; Grosso, S. J.Del ; Dickprof, W. A. ; HelmersProf, M. J. ; Herzmann, D. ; LauerProf, J. G. ; NafzigerProf, E. D. ; SawyerProf, J. E. ; ScharfProf, P. C. ; Strockprof, J. S. ; VillamilProf, M. B. / What does it take to detect a change in soil carbon stock? a regional comparison of minimum detectable difference and experiment duration in the north central united states. In: Journal of Soil and Water Conservation. 2014 ; Vol. 69, No. 6. pp. 517-531.
@article{c49abb2d18ed4c15a28c880bdadfcd50,
title = "What does it take to detect a change in soil carbon stock? a regional comparison of minimum detectable difference and experiment duration in the north central united states",
abstract = "Variability in soil organic carbon (SOC) results from natural and human processes interacting across time and space, and leads to large variation in the minimum difference in SOC that can be detected with a particular experimental design. Here we report a unique comparison of minimum detectable differences (MDDs) in SOC, and the estimated times required to observe those MDDs across the north central United States, calculated for the two most common SOC experiments: (1) a comparison between two treatments, e.g., moldboard plow (MP) and no-tillage (NT), using a randomized complete block design experiment; and (2) a comparison of changes in SOC over time for a particular treatment, e.g., NT, using a randomized complete block design experiment with time as an additional factor. We estimated the duration of the two experiment types required to achieve MDD through simulation of SOC dynamics. Data for the study came from 13 experimental sites located in Iowa, Illinois, Ohio, Michigan,Wisconsin, Missouri, and Minnesota. Soil organic carbon, bulk density, and texture were measured at four soil depths. Minimum detectable differences were calculated with probability ofType I error of 0.05 and probability ofType II error of 0.15. The MDDs in SOC were highly variable across the region and increased with soil depth. At 0 to 10 cm (0 to 3.9 in) soil depth, MDDs with five replications ranged from 1.04 g C kg-1 (0.017 oz C lb-1; 6{\%}) to 7.15 g C kg-1 (0.114 oz C lb-1; 31{\%}) for comparison of two treatments; and from 0.46 g C kg-1 (0.007 oz C lb-1; 3{\%}) to 3.12 g C kg-1 (0.050 oz C lb-1; 13{\%}) for SOC change over time. Large differences were also predicted in the experiment duration required to detect a difference in SOC between MP and NT (from 8 to >100 years with five replications), or a change in SOC over time under NT management (from 11 to 71 years with five replications). At most locations, the time required to detect a change in SOC under NT was shorter than the time required to detect a difference between MP and NT. Minimum detectable difference and experiment duration decreased with the number of replications and were correlated with SOC variability and soil texture of the experimental sites, i.e., they tended to be lower in fine textured soils. Experiment duration was also reduced by increased crop productivity and the amount of residue left on the soil. The relationships and methods described here enable the design of experiments with high power of detecting differences and changes in SOC and enhance our understanding of how management practices influence SOC storage.",
keywords = "Experiment duration, Minimum detectable difference, No-till, Soil organic carbon, Statistical power, Tillage",
author = "M. Necp{\'a}lov{\'a} and Anex, {R. P.} and KxravchenkoProf, {A. N.} and Abendroth, {L. J.} and Grosso, {S. J.Del} and Dickprof, {W. A.} and HelmersProf, {M. J.} and D. Herzmann and LauerProf, {J. G.} and NafzigerProf, {E. D.} and SawyerProf, {J. E.} and ScharfProf, {P. C.} and Strockprof, {J. S.} and VillamilProf, {M. B.}",
year = "2014",
month = "11",
day = "1",
doi = "10.2489/jswc.69.6.517",
language = "English (US)",
volume = "69",
pages = "517--531",
journal = "Journal of Soils and Water Conservation",
issn = "0022-4561",
publisher = "Soil Conservation Society of America",
number = "6",

}

TY - JOUR

T1 - What does it take to detect a change in soil carbon stock? a regional comparison of minimum detectable difference and experiment duration in the north central united states

AU - Necpálová, M.

AU - Anex, R. P.

AU - KxravchenkoProf, A. N.

AU - Abendroth, L. J.

AU - Grosso, S. J.Del

AU - Dickprof, W. A.

AU - HelmersProf, M. J.

AU - Herzmann, D.

AU - LauerProf, J. G.

AU - NafzigerProf, E. D.

AU - SawyerProf, J. E.

AU - ScharfProf, P. C.

AU - Strockprof, J. S.

AU - VillamilProf, M. B.

PY - 2014/11/1

Y1 - 2014/11/1

N2 - Variability in soil organic carbon (SOC) results from natural and human processes interacting across time and space, and leads to large variation in the minimum difference in SOC that can be detected with a particular experimental design. Here we report a unique comparison of minimum detectable differences (MDDs) in SOC, and the estimated times required to observe those MDDs across the north central United States, calculated for the two most common SOC experiments: (1) a comparison between two treatments, e.g., moldboard plow (MP) and no-tillage (NT), using a randomized complete block design experiment; and (2) a comparison of changes in SOC over time for a particular treatment, e.g., NT, using a randomized complete block design experiment with time as an additional factor. We estimated the duration of the two experiment types required to achieve MDD through simulation of SOC dynamics. Data for the study came from 13 experimental sites located in Iowa, Illinois, Ohio, Michigan,Wisconsin, Missouri, and Minnesota. Soil organic carbon, bulk density, and texture were measured at four soil depths. Minimum detectable differences were calculated with probability ofType I error of 0.05 and probability ofType II error of 0.15. The MDDs in SOC were highly variable across the region and increased with soil depth. At 0 to 10 cm (0 to 3.9 in) soil depth, MDDs with five replications ranged from 1.04 g C kg-1 (0.017 oz C lb-1; 6%) to 7.15 g C kg-1 (0.114 oz C lb-1; 31%) for comparison of two treatments; and from 0.46 g C kg-1 (0.007 oz C lb-1; 3%) to 3.12 g C kg-1 (0.050 oz C lb-1; 13%) for SOC change over time. Large differences were also predicted in the experiment duration required to detect a difference in SOC between MP and NT (from 8 to >100 years with five replications), or a change in SOC over time under NT management (from 11 to 71 years with five replications). At most locations, the time required to detect a change in SOC under NT was shorter than the time required to detect a difference between MP and NT. Minimum detectable difference and experiment duration decreased with the number of replications and were correlated with SOC variability and soil texture of the experimental sites, i.e., they tended to be lower in fine textured soils. Experiment duration was also reduced by increased crop productivity and the amount of residue left on the soil. The relationships and methods described here enable the design of experiments with high power of detecting differences and changes in SOC and enhance our understanding of how management practices influence SOC storage.

AB - Variability in soil organic carbon (SOC) results from natural and human processes interacting across time and space, and leads to large variation in the minimum difference in SOC that can be detected with a particular experimental design. Here we report a unique comparison of minimum detectable differences (MDDs) in SOC, and the estimated times required to observe those MDDs across the north central United States, calculated for the two most common SOC experiments: (1) a comparison between two treatments, e.g., moldboard plow (MP) and no-tillage (NT), using a randomized complete block design experiment; and (2) a comparison of changes in SOC over time for a particular treatment, e.g., NT, using a randomized complete block design experiment with time as an additional factor. We estimated the duration of the two experiment types required to achieve MDD through simulation of SOC dynamics. Data for the study came from 13 experimental sites located in Iowa, Illinois, Ohio, Michigan,Wisconsin, Missouri, and Minnesota. Soil organic carbon, bulk density, and texture were measured at four soil depths. Minimum detectable differences were calculated with probability ofType I error of 0.05 and probability ofType II error of 0.15. The MDDs in SOC were highly variable across the region and increased with soil depth. At 0 to 10 cm (0 to 3.9 in) soil depth, MDDs with five replications ranged from 1.04 g C kg-1 (0.017 oz C lb-1; 6%) to 7.15 g C kg-1 (0.114 oz C lb-1; 31%) for comparison of two treatments; and from 0.46 g C kg-1 (0.007 oz C lb-1; 3%) to 3.12 g C kg-1 (0.050 oz C lb-1; 13%) for SOC change over time. Large differences were also predicted in the experiment duration required to detect a difference in SOC between MP and NT (from 8 to >100 years with five replications), or a change in SOC over time under NT management (from 11 to 71 years with five replications). At most locations, the time required to detect a change in SOC under NT was shorter than the time required to detect a difference between MP and NT. Minimum detectable difference and experiment duration decreased with the number of replications and were correlated with SOC variability and soil texture of the experimental sites, i.e., they tended to be lower in fine textured soils. Experiment duration was also reduced by increased crop productivity and the amount of residue left on the soil. The relationships and methods described here enable the design of experiments with high power of detecting differences and changes in SOC and enhance our understanding of how management practices influence SOC storage.

KW - Experiment duration

KW - Minimum detectable difference

KW - No-till

KW - Soil organic carbon

KW - Statistical power

KW - Tillage

UR - http://www.scopus.com/inward/record.url?scp=84912021853&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84912021853&partnerID=8YFLogxK

U2 - 10.2489/jswc.69.6.517

DO - 10.2489/jswc.69.6.517

M3 - Article

AN - SCOPUS:84912021853

VL - 69

SP - 517

EP - 531

JO - Journal of Soils and Water Conservation

JF - Journal of Soils and Water Conservation

SN - 0022-4561

IS - 6

ER -