Training Data Selection for Annual Land Cover Classification for the Land Change Monitoring, Assessment, and Projection (LCMAP) Initiative

Heather Tollerud,Qiang Zhou,Christopher Barber,Kelcy Smith,Daniel Zelenak

doi:10.3390/rs12040699

Heather Tollerud, Qiang Zhou + Show 3 more

Open Access

https://doi.org/10.3390/rs12040699

Copy DOI

Abstract

The U.S. Geological Survey’s Land Change Monitoring, Assessment, and Projection (LCMAP) initiative involves detecting changes in land cover, use, and condition with the goal of producing land change information to improve the understanding of the Earth system and provide insights on the impacts of land surface change on society. The change detection method ingests all available high-quality data from the Landsat archive in a time series approach to identify the timing and location of land surface change. Annual thematic land cover maps are then produced by classifying time series models. In this paper, we describe the optimization of the classification method used to derive the thematic land cover product. We investigated the influences of auxiliary data, sample size, and training from different sources such as the U.S. Geological Survey’s Land Cover Trends project and National Land Cover Database (NLCD 2001 and NLCD 2011). The results were evaluated and validated based on independent data from the training dataset. We found that refining the auxiliary data effectively reduced artifacts in the thematic land cover map that are related to data availability. We improved the classification accuracy and stability considerably by using a total of 20 million training pixels with a minimum of 600,000 and a maximum of 8 million training pixels per class within geographic windows consisting of nine Analysis Ready Data tiles (450 by 450 km2). Comparisons revealed that the NLCD 2001 training data delivered the best classification accuracy. Compared to the original LCMAP classification strategy used for early evaluation (e.g., Trends training data, 20,000 samples), the optimized classification strategy improved the annual land cover map accuracy by an average of 10%.

Highlights

Land cover and land change play a major role in the climate and biogeochemistry of the Earth system [1]
From the Analysis Ready Data (ARD) data, LCMAP derives annual land cover maps using an adaptation of the Continuous Change Detection and Classification (CCDC) algorithm [13,14]
The initial CCDC classification strategy was designed based on five Landsat path/rows across the conterminous United States (CONUS), with training data from map products of the U.S Geological Survey (USGS) Land Cover Trends project [15]

Summary

Introduction

Land cover and land change play a major role in the climate and biogeochemistry of the Earth system [1]. From the ARD data, LCMAP derives annual land cover maps using an adaptation of the Continuous Change Detection and Classification (CCDC) algorithm [13,14]. The distribution and balancing of training data among classes are crucial in machine-learning classification [23]. Distribution refers to the proportion of training data for a specific class, which is often related to the population of the map classes in the study area [15]. The initial CCDC classification strategy was designed based on five Landsat path/rows across the conterminous United States (CONUS), with training data from map products of the USGS Land Cover Trends project [15]. (2) What is the optimum amount of training data? (3) What is the optimum source of training data?

Data and Study Area

LCMAP Continuous Change Detection

Auxiliary Data

Study Area

Auxiliary Data Refining

Training Sample Size Optimization

Training Data Source Optimization