Free access

Updating Canada’s National Forest Inventory with multiple imputations of missing contemporary data

Publication: The Forestry Chronicle
2 November 2017

Abstract

Canada’s National Forest Inventory (NFI) is facing an issue of spatial imbalance in photo interpreted data from 400 ha photo-plots available for estimation of state and change. Multiple imputations (MI) of missing data is therefore considered as a means to mitigate a potential bias arising from spatial imbalance, and—to a lesser degree— improve the precision relative to what can be achieved with the subset of plots having current data. In this study we explored MI with data from three study sites located in the provinces of Quebec, Ontario, and Saskatchewan. Specifically, we looked at state at time T2 and change between T1 and T2 in cover-type area proportions and in per unit area stem volume. At each location we found significant T1 differences in these attributes between plots with and without T2 data. A MI procedure with 20 replications of stochastic model-based imputations of missing data was therefore effective as a way to mitigate a bias that would arise if T2 inference was based exclusively on plots with T2 data. Possible differences between the T2 and T1 photointerpretation, paired with no efficient stratification of disturbed and undisturbed plots, largely eliminated expected gains in precision from the MI boosting of the effective T2 sample size. Despite recognized limitations, we recommend MI as an effective tool to counteract an emerging spatial imbalance in the NFI.

Résumé

L’inventaire forestier national du Canada (NFI) est confronté à une problématique de déséquilibre dans les données de placettes photographiques de 400 ha disponibles pour évaluer l’état de la forêt et son changement. On envisage donc d’utiliser l’imputation multiple (MI) des données manquantes afin d’amoindrir le biais occasionné par le déséquilibre spatial et, dans une moindre mesure, améliorer la précision par rapport à ce qu’il est possible d’atteindre avec le sous-ensemble de placettes pour lesquelles nous disposons de données à jour. Dans cette étude, nous avons exploré les possibilités de MI avec les données provenant de trois sites d’étude situés dans les provinces de Québec, de l’Ontario et de la Saskatchewan. De façon plus particulière, nous nous sommes intéressés à l’état à temps T2 et au changement survenu entre T1 et T2 dans les proportions des surfaces par type de couvert et le volume par tige par unité de surface. À chaque endroit nous avons observé des différences significatives pour ces paramètres à T1 selon qu’on utilisait ou non les données de T2. Une procédure MI avec 20 répétitions d’imputations des données manquantes basées sur un modèle stochastique s’est avérée efficace pour atténuer le biais qui résulterait si l’inférence à T2 reposait uniquement sur les placettes disposant des données à T2. Toutefois, les différences dans la technique de photo-interprétation à T1 et à T2 combinées à l’absence de stratification des placettes perturbées et intactes a masqué la majeure partie du gain en précision qu’on aurait pu espérer avec une bonification de la taille des échantillons à T2. Malgré ces limitations, nous recommandons l’utilisation de MI pour amoindrir les effets d’un déséquilibre anticipé dans l’inventaire forestier national.

Formats available

You can view the full content in the following formats:

References

Bokalo M., Titus S.J., and Wiens D.P. 1996 Sampling with partial replacement extended to include growth projections For. Sci 42 328 -334
Brown, C.H. 1990. Protecting against nonrandomly missing data in longitudinal studies. Biometrics.
Carpenter, J.R. and M.G. Kenward. 2013.Multiple imputation and its application. Wiley, 345 p.
Carroll, R.J., D. Ruppert and L.A. Stefanski. 1995. Measurement error in nonlinear models. Chapman & Hall, London, 305 p.
Chen J.H. and Shao J. 2001 Jackknife variance estimation for nearest-neighbor imputation J. Am. Stat. Assoc 96 453 260 -269
Cochran, W.G. 1977. Sampling techniques. Wiley, New York, 380 p.
Corona P., Chirici G., Franceschi S., Maffei D., Marcheselli M., Pisani C., et al. 2014 Design-based treatment of missing data in forest inventories using canopy heights from aerial laser scanning Can. J. For. Res 44 8 892 -902
Crookston N.L. and Finley A.O. 2008 yaImpute: An R Package for kNN Imputation J. Stat. Softw 23 10 16
Cunia T. and Kyaw U.S. 1985 Sampling with partial replacement with different types of samples Can. J. For. Res 15 191 -194
Dixon B.L. and Howitt R.E. 1979 Continuous Forest Inventory Using a Linear Filter For. Sci 25 4 675 -689
Domke G.M., Woodall C.W., Walters B.F., McRoberts R.E., and Hatfield M.A. 2014 Strategies to compensate for the effects of non-response on forest carbon baseline estimates from the national forest inventory of the United States For. Ecol. Manage 315 112 -120
Eskelson B.N.I., Temesgen H., and Barrett T.M. 2009a Estimating Current Forest Attributes from Paneled Inventory Data Using Plot-Level Imputation: A Study from the Pacific Northwest For. Sci 55 1 64 -71
Eskelson B.N.I., Temesgen H., Lemay V., Barrett T.M., Crookston N.L., and Hudak A.T. 2009b The roles of nearest neighbor methods in imputing missing data in forest inventory and monitoring databases Scand. J. For. Res 24 3 235 -246
Fuller, W.A. 1987.Measurement Error Models. Wiley, New York.
Fuller, W.A. 2009. Sampling Statistics. Wiley, New York, 454 p.
Gillis M.D., Omule A.Y., and Brierley T. 2005 Monitoring Canada’s forests: The National Forest Inventory For. Chron 81 2 214 -221
Glynn R.J., Laird N.M., and Rubin D.B. 1993 Multiple imputation in mixture models for nonignorable nonresponse with follow-up J. Am. Stat. Assoc 88 984 -993
Good, P. 1993. Permutation tests. Springer, New York, 226 p.
Greene, W.H. 2012. Econometric analysis. 7 edn. Prentice Hall, Upper Saddle River, NJ.
Gregoire T.G. and Walters D.K. 1988 Composite vector estimator by weighting inversely proportional to variance Can. J. For. Res 18 282 -284
Haywood A., Mellor A., and Stone C. 2016 A strategic forest inventory for public land in Victoria, Australia For. Ecol. Manage 367 86 -96 https://doi.org/10.1016/j.foreco.2016.02.026
Hermosilla T., Wulder M.A., White J.C., Coops N.C., and Hobart G.W. 2015 An integrated Landsat time series protocol for change detection and generation of annual gap-free surface reflectance composites Remote. Sens. Environ 158 220 -234
Holmström H. 2002 Estimation of single-tree characteristics using the kNN method and plotwise aerial photograph interpretations For. Ecol. Manage 167 1–3 303 -314
Holmström H., Nilsson M., and Ståhl G. 2001 Simultaneous estimations of forest parameters using aerial photograph interpreted data and the k nearest neighbour method Scand. J. For. Res 16 1 67 -78
Jirousek R. and Preucil S. 1995 On the effective implementation of the iterative proportional fitting procedure Comp. Stat. Data Anal 19 2 177 -189
Kangas A. 1991 Updated measurement data as prior information in forest inventory Silv. Fenn 25 3 10
Kim J.K. and Fuller W. 2004 Fractional hot deck imputation Biometrika 91 3 559 -578
Koehler E., Brown E., and Haneuse J.-P.A. 2009 On the assessment of Monte Carlo error in simulation-based statistical analyses Am. Stat 63 2 155 -162
Little, R.J.A. and D.B. Rubin. 1987. Statistical analysis with missing data. Wiley.
MacDicken K.G. 2015 Global forest resources assessment 2015: What, why and how? For Ecol. Manage 352 3 -8
Magnussen S. and Russo G. 2012 Uncertainty in photo-interpreted forest inventory variables and effects on estimates of error in Canada’s National Forest Inventory Forest. Chron 88 4 439 -447
Marimont R.B. and Shapiro M.B. 1979 Nearest Neighbour Searches and the Curse of Dimensionality IMA J. Appl. Math 24 1 59 -70
Massey A., Mandallaz D., and Lanz A. 2014 Integrating remote sensing and past inventory data under the new annual design of the Swiss National Forest Inventory using three-phase design-based regression estimation Can. J. For. Res 44 1177 -1186
Matis K.G., Hetherington J.C., and Kassab J.Y. 1984 Sampling with partial replacement - A literature review Comm. For. Rev 63 193 -206
McRoberts R.E. 2001 Imputation and model-based updating techniques for annual forest inventories For. Sci 47 3 322 -330
McRoberts R.E. 2003 Compensating for missing plot observations in forest inventory estimation Can. J. For. Res 33 10 1990 -1997
McRoberts R.E., Næsset E., and Gobakken T. 2016 The effects of temporal differences between map and ground data on map-assisted estimates of forest area and biomass Ann. For. Sci 73 4 839 -847
Miller, R.G.J. 1981. Simultaneous Statistical Inference. 2 edn. Springer, New York, 293 p.
Patterson P.L., Coulston J.W., Roesch F.A., Westfall J.A., and Hill A.D. 2012 A primer for nonresponse in the US forest inventory and analysis program Environ. Monit. Assess 184 3 1423 -1433
Pelz, D.R. and T. Cunia. 1985. National Forest Inventories in Europe. Univ. of Freiburg, Freiburg, Germany.
Reams G.A. and Van Deusen P.C. 1999 The southern annual forest inventory system J. Agric. Biol. Environ. Stat 4 4 346 -360
Reams, G.A. and J.M. McCollum. 2000. The use of multiple imputation in the southern annual forest inventory system. Gen. Tech. Rep, No. NC-122, US Forest Service, Moscow ID.
Rubin, D.B. 1987. Multiple imputation for nonresponse in surveys. Wiley, New York, 261 p.
Scott C.T. 1984 A new look at sampling with partial replacement For. Sci 30 157 -166
Ståhl G., Allard A., Esseen P.A., Glimskär A., Ringvall A., Svensson J., et al. 2011 National Inventory of Landscapes in Sweden (NILS)-scope, design, and experiences from establishing a multiscale biodiversity monitoring system Env. Monit. Assess 173 1-4 579 -595
StataCorp. 2016. Stata Statistical Software: Release 14.2. StataCorp LP, College Station, Texas.
Stinson, G., S. Magnussen, P. Boudewyn, F. Eichel, G. Russo, M. Cranny et al. 2016. Canada. In: National Forest Inventories: Assessment of Wood Availability and Use. C. Vidal, I.A. Alberdi, L.H. Mateo and J.J. Redmond (eds). Springer, Cham, CH., pp. 233–248.
Tang G., Little R.J.A., and Raghunathan T.E. 2003 Analysis of multivariate missing data with nonignorable nonresponse Biometrika 90 4 747 -764
Tomppo, E. 2006. The Finnish National Forest Inventory. In: Forest Inventory: Methodology and Applications. A. Kangas and M. Maltamo (eds). Springer, Dordrecht, pp. 179–194.
Traub B., Meile R., Speich S., and Rösler E. 2017 The data storage and analysis system of the Swiss National Forest Inventory Computers and Electronics in Agriculture 132 97 -107
Van Deusen P.C. 1996 Incorporating predictions into an annual forest inventory Can. J. For. Res 26 1709 -1713
Van Deusen P.C. 1997 Annual forest inventory statistical concepts with emphasis on multiple imputations Can. J. For. Res 27 379 -384
White J.C., Wulder M.A., Hobart G.W., Luther J.E., Hermosilla T., Griffiths P., et al. 2014 Pixel-based image compositing for large-area dense time series applications and science Can. J. Rem. Sens 40 3
Wolter, K.M. 2007. Introduction to variance estimation. 2nd edn. Springer, New York, 447 p.
Wulder, M.A. and T. Nelson. 2003. EOSD land cover classification legend report., Victoria British Columbia, pp. 1–81.
Wulder M.A., White J.C., Cranny M., Hall R.J., Luther J.E., Beaudoin A., et al. 2008 Monitoring Canada’s Forests. Part 1: Completion of the EOSD Land Cover Project Can. J. Rem. Sens 34 6 549 -562

Information & Authors

Information

Published In

cover image The Forestry Chronicle
The Forestry Chronicle
Volume 93Number 03October 2017
Pages: 213 - 225

History

Version of record online: 2 November 2017

Key Words

  1. forest cover-types
  2. area proportions
  3. stem volume
  4. disturbance rates
  5. ratio estimators

Mots-clés

  1. types de couverts forestiers
  2. proportion des surfaces
  3. volume par tige
  4. taux de perturbation
  5. estimateurs par quotients

Authors

Affiliations

Steen Magnussen [email protected]
Canadian Forest Service, 506 West Burnside Road, Victoria BC V8Z 1M5, Canada
Graham Stinson
Canadian Forest Service, 506 West Burnside Road, Victoria BC V8Z 1M5, Canada
Paul Boudewyn
Canadian Forest Service, 506 West Burnside Road, Victoria BC V8Z 1M5, Canada

Metrics & Citations

Metrics

Other Metrics

Citations

Cite As

Export Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

Cited by

1. Take five: about the beat and the bar of annual and 5-year periodic national forest inventories
2. Relationships between population characteristics and nonresponse in urban forest inventories
3. Addressing nonresponse bias in forest inventory change estimation using response homogeneity classifications
4. An estimation method to reduce complete and partial nonresponse bias in forest inventory
5. A spatial and temporal assessment of nonresponse in the national forest inventory of the U.S
6. Nonresponse bias in change estimation: a national forest inventory example
7. Advancing the application of remote sensing for forest information needs in Canada: Lessons learned from a national collaboration of university, industrial and government stakeholders
8. Progrès dans l’application de la télédétection pour les besoins en matière d’information sur les forêts au Canada : leçons tirées d’une collaboration nationale d’intervenants universitaires, industriels et gouvernementaux

View Options

View options

PDF

View PDF

Media

Tables

Media

Share Options

Share

Share the article link

Share on social media