Journal Home > Volume 11 , Issue 1
Forest Ecosystems 2024, 11(1): 100164 https://doi.org/10.1016/j.fecs.2023.100164
Research Article | Open Access | Issue | Published: 03 January 2024

Why ecosystem characteristics predicted from remotely sensed data are unbiased and biased at the same time – and how this affects applications

Show Author's Information Göran Ståhla( ),  Terje Gobakkenb,  Svetlana Saarelab,  Henrik J. Perssona,  Magnus Ekströma,  Sean P. Healeyc,  Zhiqiang Yangc,  Johan Holmgrena,  Eva Lindberga,  Kenneth Nyströma,  Emanuele Papuccia,  Patrik Ulvdala,  Hans Ole Ørkab,  Erik Næssetb,  Zhengyang Houd,  Håkan Olssona,  Ronald E. McRobertse
Department of Forest Resource Management, Swedish University of Agricultural Sciences, 901 83, Umeå, Sweden
Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences, Ås, Norway
USDA Forest Service, Rocky Mountain Research Station, Ogden, UT, USA
The Key Laboratory for Silviculture and Conservation of Ministry of Education, Beijing Forestry University, Beijing, 100083, China
Department of Forest Resources, University of Minnesota, St. Paul, MN, USA
Keywords:
Bias, Design-based inference, Model-based inference
Cite this article:
Ståhl G, Gobakken T, Saarela S, et al. Why ecosystem characteristics predicted from remotely sensed data are unbiased and biased at the same time – and how this affects applications. Forest Ecosystems, 2024, 11(1): 100164. https://doi.org/10.1016/j.fecs.2023.100164
Download citation
EndNote(RIS)
BibTeX

9

Views

1

Downloads

Citations

0

Crossref

1

WoS

0

Scopus

0

CSCD

Abstract Full text About this article

Remotely sensed data are frequently used for predicting and mapping ecosystem characteristics, and spatially explicit wall-to-wall information is sometimes proposed as the best possible source of information for decision-making. However, wall-to-wall information typically relies on model-based prediction, and several features of model-based prediction should be understood before extensively relying on this type of information. One such feature is that model-based predictors can be considered both unbiased and biased at the same time, which has important implications in several areas of application. In this discussion paper, we first describe the conventional model-unbiasedness paradigm that underpins most prediction techniques using remotely sensed (or other) auxiliary data. From this point of view, model-based predictors are typically unbiased. Secondly, we show that for specific domains, identified based on their true values, the same model-based predictors can be considered biased, and sometimes severely so.

We suggest distinguishing between conventional model-bias, defined in the statistical literature as the difference between the expected value of a predictor and the expected value of the quantity being predicted, and design-bias of model-based estimators, defined as the difference between the expected value of a model-based estimator and the true value of the quantity being predicted. We show that model-based estimators (or predictors) are typically design-biased, and that there is a trend in the design-bias from overestimating small true values to underestimating large true values. Further, we give examples of applications where this is important to acknowledge and to potentially make adjustments to correct for the design-bias trend. We argue that relying entirely on conventional model-unbiasedness may lead to mistakes in several areas of application that use predictions from remotely sensed data.


menu
Abstract
Full text
Outline
About this article

Why ecosystem characteristics predicted from remotely sensed data are unbiased and biased at the same time – and how this affects applications

Show Author's information Göran Ståhla( ), Terje Gobakkenb, Svetlana Saarelab, Henrik J. Perssona, Magnus Ekströma, Sean P. Healeyc, Zhiqiang Yangc, Johan Holmgrena, Eva Lindberga, Kenneth Nyströma, Emanuele Papuccia, Patrik Ulvdala, Hans Ole Ørkab, Erik Næssetb, Zhengyang Houd, Håkan Olssona, Ronald E. McRobertse
Department of Forest Resource Management, Swedish University of Agricultural Sciences, 901 83, Umeå, Sweden
Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences, Ås, Norway
USDA Forest Service, Rocky Mountain Research Station, Ogden, UT, USA
The Key Laboratory for Silviculture and Conservation of Ministry of Education, Beijing Forestry University, Beijing, 100083, China
Department of Forest Resources, University of Minnesota, St. Paul, MN, USA

Abstract

Remotely sensed data are frequently used for predicting and mapping ecosystem characteristics, and spatially explicit wall-to-wall information is sometimes proposed as the best possible source of information for decision-making. However, wall-to-wall information typically relies on model-based prediction, and several features of model-based prediction should be understood before extensively relying on this type of information. One such feature is that model-based predictors can be considered both unbiased and biased at the same time, which has important implications in several areas of application. In this discussion paper, we first describe the conventional model-unbiasedness paradigm that underpins most prediction techniques using remotely sensed (or other) auxiliary data. From this point of view, model-based predictors are typically unbiased. Secondly, we show that for specific domains, identified based on their true values, the same model-based predictors can be considered biased, and sometimes severely so.

We suggest distinguishing between conventional model-bias, defined in the statistical literature as the difference between the expected value of a predictor and the expected value of the quantity being predicted, and design-bias of model-based estimators, defined as the difference between the expected value of a model-based estimator and the true value of the quantity being predicted. We show that model-based estimators (or predictors) are typically design-biased, and that there is a trend in the design-bias from overestimating small true values to underestimating large true values. Further, we give examples of applications where this is important to acknowledge and to potentially make adjustments to correct for the design-bias trend. We argue that relying entirely on conventional model-unbiasedness may lead to mistakes in several areas of application that use predictions from remotely sensed data.

Keywords: Bias, Design-based inference, Model-based inference

References(47)

Arnab, R., 2017. Survey Sampling Theory and Applications. Academic Press.
DOI

Andersen, H.E., Strunk, J., Temesgen, H., 2011. Using airborne light detection and ranging as a sampling tool for estimating forest biomass resources in the Upper Tanana Valley of Interior Alaska. West. J. Appl. Finance 26 (4), 157–164.
DOI Google Scholar

Barnett, A.G., Van Der Pols, J.C., Dobson, A.J., 2005. Regression to the mean: what it is and how to deal with it. Int. J. Epidemiol. 34 (1), 215–220.
Google Scholar

Barth, A., Lind, T., Ståhl, G., 2012. Restricted imputation for improving spatial consistency in landscape level data for forest scenario analysis. For. Ecol. Manag. 272, 61–68.
DOI Google Scholar

Breidenbach, J., Astrup, R., 2012. Small area estimation of forest attributes in the Norwegian National Forest Inventory. Eur. J. For. Res. 131, 1255–1267.
DOI Google Scholar

Brewer, K., 2013. Three controversies in the history of survey sampling. Surv. Methodol. 39 (2), 249–263.
Google Scholar
Cassel, C.M., Särndal, C.E., Wretman, J.H., 1977. Foundations of Inference in Survey Sampling. Wiley, New York.

Chambers, R., Clark, R., 2012. An Introduction to Model-Based Survey Sampling with Applications. Oxford University Press, New York.
DOI

Dubayah, R., Armston, J., Healey, S.P., Bruening, J.M., Patterson, P.L., Kellner, J.R., Duncanson, L., Saarela, S., Ståhl, G., Yang, Z., Tang, H., Blair, J.B., Fatoyinbo, L., Goetz, S., Hancock, S., Hansen, M., Hofton, M., Luthcke, S., 2022. GEDI launches a new era of biomass inference from space. Environ. Res. Lett. 17 (9), 095001.
DOI Google Scholar

Dumelle, M., Higham, M., Ver Hoef, J.M., Olsen, A.R., Madsen, L., 2022. A comparison of design-based and model-based approaches for finite population spatial sampling and inference. Methods Ecol. Evol. 13 (9), 2018–2029.
DOI Google Scholar

Ehlers, S., Grafström, A., Nyström, K., Olsson, H., Ståhl, G., 2013. Data assimilation in stand-level forest inventories. Can. J. For. Res. 43 (12), 1104–1113.
DOI Google Scholar

Ehlers, S., Saarela, S., Lindgren, N., Lindberg, E., Nyström, M., Persson, H.J., Olsson, H., Ståhl, G., 2018. Assessing error correlations in remote sensing-based estimates of forest attributes for improved composite estimation. Rem. Sens. 10 (5), 667.
DOI Google Scholar

Galton, F., 1886. Regression towards mediocrity in hereditary stature. J. Anthropol. Inst. G. B. Ireland 15, 246–263.
DOI Google Scholar

Gao, Y., Lu, D., Li, G., Wang, G., Chen, Q., Liu, L., Li, D., 2018. Comparative analysis of modeling algorithms for forest aboveground biomass estimation in a subtropical region. Rem. Sens. 10 (4), 627.
DOI Google Scholar

Gilichinsky, M., Heiskanen, J., Barth, A., Wallerman, J., Egberth, M., Nilsson, M., 2012. Histogram matching for the calibration of k NN stem volume estimates. Int. J. Rem. Sens. 33 (22), 7117–7131.
DOI Google Scholar

Gregoire, T.G., 1998. Design-based and model-based inference in survey sampling: appreciating the difference. Can. J. For. Res. 28 (10), 1429–1447.
DOI Google Scholar

Gregoire, T.G., Ståhl, G., Næsset, E., Gobakken, T., Nelson, R., Holm, S., 2011. Model-assisted estimation of biomass in a LiDAR sample survey in Hedmark County, Norway. Can. J. For. Res. 41 (1), 83–95.
DOI Google Scholar

Hansen, M.C., Potapov, P.V., Moore, R., Hancher, M., Turubanova, S.A., Tyukavina, A., Thau, D., Stehman, S.V., Goetz, S.J., Loveland, T.R., Kommareddy, A., Egorov, A., Chini, L., Justice, C.O., Townshend, J., 2013. High-resolution global maps of 21st-century forest cover change. Science 342 (6160), 850–853.
DOI Google Scholar
Hao, L., Naiman, D.Q., 2007. Quantile Regression (No. 149). Sage, London.
DOI

Heckel, K., Urban, M., Schratz, P., Mahecha, M.D., Schmullius, C., 2020. Predicting forest cover in distinct ecosystems: the potential of multi-source Sentinel-1 and-2 data fusion. Rem. Sens. 12 (2), 302.
DOI Google Scholar

Heeringa, S.G., West, B.T., Berglund, P.A., 2017. Applied Survey Data Analysis. CRC press. https://doi.org/10.1201/9781315153278.
DOI

Hou, Z., Mehtätalo, L., McRoberts, R.E., Ståhl, G., Tokola, T., Rana, P., Siipilehto, J., Xu, Q., 2019. Remote sensing-assisted data assimilation and simultaneous inference for forest inventory. Remote Sens. Environ. 234, 111431.
DOI Google Scholar

Hou, Z., McRoberts, R.E., Zhang, C., Ståhl, G., Zhao, X., Wang, X., Li, B., Xu, Q., 2022. Cross-classes domain inference with network sampling for natural resource inventory. For. Ecosyst. 9, 100029.
DOI Google Scholar

Langner, A., Achard, F., Grassi, G., 2014. Can recent pan-tropical biomass maps be used to derive alternative Tier 1 values for reporting REDD+ activities under UNFCCC? Environ. Res. Lett. 9 (12), 124008.
DOI Google Scholar

Lindgren, N., Nyström, K., Saarela, S., Olsson, H., Ståhl, G., 2022. Importance of calibration for improving the efficiency of data assimilation for predicting forest characteristics. Rem. Sens. 14 (18), 4627.
DOI Google Scholar

Lämås, T., Sängstuvall, L., Öhman, K., Lundström, J., Årevall, J., Holmström, H., Nilsson, L., Nordstrom, E., Wikberg, P., Wikstrom, P., Eggers, J., 2023. The multi-faceted Swedish Heureka forest decision support system: context, functionality, design, and 10 years experiences of its use. Front. For. Glob. Change 6, 1163105.
DOI Google Scholar

McRoberts, R.E., Næsset, E., Gobakken, T., Bollandsås, O.M., 2015. Indirect and direct estimation of forest biomass change using forest inventory and airborne laser scanning data. Remote Sens. Environ. 164, 36–42.
DOI Google Scholar

McRoberts, R.E., Næsset, E., Gobakken, T., Chirici, G., Condés, S., Hou, Z., Saarela, S., Chen, Q., Ståhl, G., Walters, B.F., 2018. Assessing components of the model-based mean square error estimator for remote sensing assisted forest applications. Can. J. For. Res. 48 (6), 642–649.
DOI Google Scholar

Mitchard, E.T., Saatchi, S.S., Baccini, A., Asner, G.P., Goetz, S.J., Harris, N.L., Brown, S., 2013. Uncertainty in the spatial distribution of tropical forest biomass: a comparison of pan-tropical maps. Carbon Bal. Manag. 8, 1–13.
DOI Google Scholar

Mohren, G.M.J., 2003. Large-scale scenario analysis in forest ecology and forest management. For. Pol. Econ. 5 (2), 103–110.
DOI Google Scholar

Næsset, E., Gobakken, T., Solberg, S., Gregoire, T.G., Nelson, R., Ståhl, G., Weydahl, D., 2011. Model-assisted regional forest biomass estimation using LiDAR and InSAR as auxiliary data: a case study from a boreal forest area. Remote Sens. Environ. 115 (12), 3599–3614.
DOI Google Scholar

Penner, M., Pitt, D.G., Woods, M.E., 2013. Parametric vs. nonparametric LiDAR models for operational forest inventory in boreal Ontario. Can. J. Rem. Sens. 39 (5), 426–443.
Google Scholar

Persson, H.J., Ståhl, G., 2020. Characterizing uncertainty in forest remote sensing studies. Rem. Sens. 12 (3), 505.
DOI Google Scholar

Quah, D., 1993. Galton's fallacy and tests of the convergence hypothesis. Scand. J. Econ., 427–443.
DOI Google Scholar

Réjou-Méchain, M., Barbier, N., Couteron, P., Ploton, P., Vincent, G., Herold, M., Mermoz, S., Saatchi, S., Chave, J., de Boissieu, F., Feret, J., Takoudjou, S.M., Pélissier, R., 2019. Upscaling forest biomass from field to satellite measurements: sources of errors and ways to reduce them. Surv. Geophys. 40, 881–911.
DOI Google Scholar

Ruckelshaus, M., Hartway, C., Kareiva, P., 1997. Assessing the data requirements of spatially explicit dispersal models. Conserv. Biol. 11 (6), 1298–1306.
DOI Google Scholar

Saarela, S., Varvia, P., Korhonen, L., Yang, Z., Patterson, P.L., Gobakken, T., Næsset, E., Healey, S.P., Ståhl, G., 2023. Three-phase hierarchical model-based and hybrid inference. MethodsX 11, 102321.
DOI Google Scholar

Solberg, S., Weydahl, D.J., Nasset, E., 2010. Simulating X-band interferometric height in a spruce forest from airborne laser scanning. IEEE Trans. Geosci. Rem. Sens. 48 (9), 3369–3378.
DOI Google Scholar
Särndal, C.E., Swensson, B., Wretman, J., 2003. Model Assisted Survey Sampling. Springer, Berlin, Heidelberg.

Shukla, G.K., 1972. On the problem of calibration. Technometrics 14 (3), 547–553.
DOI Google Scholar

Tellinghuisen, J., 2000. Inverse vs. classical calibration for small data sets. Fresen. J. Anal. Chem. 368, 585–588.
DOI Google Scholar

Tian, Y., Nearing, G.S., Peters-Lidard, C.D., Harrison, K.W., Tang, L., 2016. Performance metrics, error modeling, and uncertainty quantification. Mon. Weather Rev. 144 (2), 607–613.
DOI Google Scholar

Tomppo, E., Olsson, H., Ståhl, G., Nilsson, M., Hagner, O., Katila, M., 2008. Combining national forest inventory field plots and remote sensing data for forest databases. Remote Sens. Environ. 112 (5), 1982–1999.
DOI Google Scholar
Thompson, S.K., 2012. Sampling, vol. 755. John Wiley & Sons, New Jersey.

Wang, R., Chen, J.M., Liu, Z., Arain, A., 2017. Evaluation of seasonal variations of remotely sensed leaf area index over five evergreen coniferous forests. ISPRS J. Photogrammetry Remote Sens. 130, 187–201.
DOI Google Scholar

Wilhelmsson, P., Sjödin, E., Wästlund, A., Wallerman, J., Lämås, T., Öhman, K., 2021. Dynamic treatment units in forest planning using cell proximity. Can. J. For. Res. 51 (7), 1065–1071.
DOI Google Scholar
Wu, C., Thompson, M.E., 2020. Sampling Theory and Practice. Springer, Cham.
DOI
Publication history
Copyright
Acknowledgements
Rights and permissions

Publication history

Received: 02 August 2023
Revised: 21 November 2023
Accepted: 23 December 2023
Published: 03 January 2024
Issue date: February 2024

Copyright

© 2024 The Authors.

Acknowledgements

This work is part of the programme Mistra Digital Forests and of the Center for Research-based Innovation SmartForest: Bringing Industry 4.0 to the Norwegian forest sector (NFR SFI project no. 309671, smartforest.no).

Rights and permissions

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Return