Making better use of accuracy data in land change studies: Estimating accuracy and area and quantifying uncertainty using stratified estimation Pontus Olofsson a, ⁎, Giles M. Foody b , Stephen V. Stehman c , Curtis E. Woodcock a a Department of Earth and Environment, Boston University, 675 Commonwealth Avenue, Boston, MA 02215, USA b School of Geography, University of Nottingham, University Park, Nottingham NG7 2RD, UK c Department of Forest and Natural Resources Management, State University of New York, 1 Forestry Drive, Syracuse, NY 13210, USA abstract article info Article history: Received 15 February 2012 Received in revised form 23 October 2012 Accepted 26 October 2012 Available online 29 November 2012 Keywords: Land use change Land cover change Carbon modeling Uncertainty Stratified estimation Accuracy assessment The area of land use or land cover change obtained directly from a map may differ greatly from the true area of change because of map classification error. An error-adjusted estimator of area can be easily produced once an accuracy assessment has been performed and an error matrix constructed. The estimator presented is a stratified estimator which is applicable to data acquired using popular sampling designs such as stratified random, simple random and systematic (the stratified estimator is often labeled a poststratified estimator for the latter two designs). A confidence interval for the area of land change should also be provided to quan- tify the uncertainty of the change area estimate. The uncertainty of the change area estimate, as expressed via the confidence interval, can then subsequently be incorporated into an uncertainty analysis for applications using land change area as an input (e.g., a carbon flux model). Accuracy assessments published for land change studies should report the information required to produce the stratified estimator of change area and to construct confidence intervals. However, an evaluation of land change articles published between 2005 and 2010 in two remote sensing journals revealed that accuracy assessments often fail to include this key information. We recommend that land change maps should be accompanied by an accuracy assessment that includes a clear description of the sampling design (including sample size and, if relevant, details of stratification), an error matrix, the area or proportion of area of each category according to the map, and de- scriptive accuracy measures such as user's, producer's and overall accuracy. Furthermore, mapped areas should be adjusted to eliminate bias attributable to map classification error and these error-adjusted area es- timates should be accompanied by confidence intervals to quantify the sampling variability of the estimated area. Using data from the published literature, we illustrate how to produce error-adjusted point estimates and confidence intervals of land change areas. A simple analysis of uncertainty based on the confidence bounds for land change area is applied to a carbon flux model to illustrate numerically that variability in the land change area estimate can have a dramatic effect on model outputs. © 2012 Elsevier Inc. All rights reserved. 1. Introduction Land use or land cover change (referred to as “land change” for the reminder of the article) impacts on a very diverse array of environmen- tal properties and processes. The effects of a land change may be felt across a broad spectrum of environmental systems including the atmo- spheric, hydrologic, geomorphologic and ecologic. Deforestation may, for example, act as a source of carbon to the atmosphere, lead to en- hanced soil erosion, reduce the extent of habitat and so to species de- clines and contribute to displacement of human populations. Land change is, therefore a critical variable in relation to two environmental issues of great societal concern: climate change and biodiversity loss. Land change can be a cause and a consequence of climate change and is a variable of greater impact than climate change (Skole, 1994). Land change is, for example, the single most important variable affecting eco- logical systems (Chapin et al., 2000; Vitousek, 1994) and the greatest threat to biodiversity (Sala et al., 2000). The importance of land change is evident in the growth of interest in land change science (Turner et al., 2007) and so there is consequently considerable interest in land cover and a need for accurate information on land cover and its dynamics. Indeed the central role of land surface change to a vast array of contem- porary concerns is reflected in its role as an underpinning feature of the current grand challenges for the geographical sciences articulated recently by the US National Academy of Sciences (CSDGSND, 2010). Remote sensing has the potential to provide accurate information on land cover but numerous problems may be encountered and the adequacy of this information has been questioned (Townshend et al., 1992; Wilkinson, 1996, 2005). In many applications the main focus is the area of a land cover class or its gross change over time (gross change refers to the total area of Remote Sensing of Environment 129 (2013) 122–131 ⁎ Corresponding author. Tel.: +1 6173539374. E-mail address: olofsson@bu.edu (P. Olofsson). URL: http://www.bu.edu/geography (P. Olofsson). 0034-4257/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.rse.2012.10.031 Contents lists available at SciVerse ScienceDirect Remote Sensing of Environment journal homepage: www.elsevier.com/locate/rse