Spatio-temporal variations in plantation forests’ disturbance and recovery of northern Guangdong Province using yearly Landsat time series observations (1986–2015)

Forest disturbance plays a vital role in modulating carbon storage, biodiversity and climate change. Yearly Landsat imagery from 1986 to 2015 of a typical plantation region in the northern Guangdong province of southern China was used as a case study. A Landsat time series stack (LTSS) was fed to the vegetation change tracker model (VCT) to map long-term changes in plantation forests’ disturbance and recovery, followed by an intensive validation and a continuous 27-yr change analysis on disturbance locations, magnitudes and rates of plantations’ disturbance and recovery. And the validation results of the disturbance year maps derived from five randomly identified sample plots with 25 km2 located at the four corners and the center of the scene showed the majority of the spatial agreement measures ranged from 60% to 83%. A confusion matrix summary of the accuracy measures for all four validation sites in Fogang County showed that the disturbance year maps had an overall accuracy estimate of 71.70%. Forest disturbance rates’ change trend was characterized by a decline first, followed by an increase, then giving way to a decline again. An undulated and gentle decreasing trend of disturbance rates from the highest value of 3.95% to the lowest value of 0.76% occurred between 1988 and 2001, disturbance rate of 4.51% in 1994 was a notable anomaly, while after 2001 there was a sharp ascending change, forest disturbance rate spiked in 2007 (5.84%). After that, there was a significant decreasing trend up to the lowest value of 1.96% in 2011 and a slight ascending trend from 2011 to 2015 (2.59%). Two obvious spikes in post-disturbance recovery rates occurred in 1995 (0.26%) and 2008 (0.41%). Overall, forest recovery rates were lower than forest disturbance rates. Moreover, forest disturbance and recovery detection based on VCT and the Landsat-based detections of trends in disturbance and recovery (LandTrendr) algorithms in Fogang County have been conducted, with LandTrendr finding mostly much more disturbance than VCT. Overall, disturbances and recoveries in northern Guangdong were triggered mostly by timber needs, policies and decisions of the local governments. This study highlights that a better understanding about plantations’ changes would provide a critical foundation for local forest management decisions in the southern China.

[1]  Zhiqiang Yang,et al.  Detecting trends in forest disturbance and recovery using yearly Landsat time series: 1. LandTrendr — Temporal segmentation algorithms , 2010 .

[2]  Christopher Justice,et al.  The impact of misregistration on change detection , 1992, IEEE Trans. Geosci. Remote. Sens..

[3]  W. Cohen,et al.  Adding value to the FIA inventory: combining FIA data and satellite observations to estimate forest disturbance , 2012 .

[4]  沈文娟 Shen Wenjuan,et al.  Mapping disturbance and recovery of plantation forests in southern China using yearly Landsat time series observations , 2017 .

[5]  Shenggong Li,et al.  Carbon dynamics and stability between native Masson pine and exotic slash pine plantations in subtropical China , 2013, European Journal of Forest Research.

[6]  Philippe Ciais,et al.  The carbon balance of terrestrial ecosystems in China , 2009, Nature.

[7]  Fengsong Pei,et al.  Exploring the response of net primary productivity variations to urban expansion and climate change: a scenario analysis for Guangdong Province in China. , 2015, Journal of environmental management.

[8]  Chengquan Huang,et al.  Use of Vegetation Change Tracker and Support Vector Machine to Map Disturbance Types in Greater Yellowstone Ecosystems in a 1984–2010 Landsat Time Series , 2015, IEEE Geoscience and Remote Sensing Letters.

[9]  Chengquan Huang,et al.  Forest disturbance across the conterminous United States from 1985-2012: The emerging dominance of forest decline , 2016 .

[10]  Scott L. Powell,et al.  Validation of North American Forest Disturbance dynamics derived from Landsat time series stacks , 2011 .

[11]  D. James.  Continuous monitoring. , 1979, The Urologic clinics of North America.

[12]  D. Edwards,et al.  Maintaining ecosystem function and services in logged tropical forests. , 2014, Trends in ecology & evolution.

[13]  S. Frolking,et al.  Forest disturbance and recovery: A general review in the context of spaceborne remote sensing of impacts on aboveground biomass and canopy structure , 2009 .

[14]  Martha C. Anderson,et al.  Free Access to Landsat Imagery , 2008, Science.

[15]  Fan Jing-yun,et al.  Dynamic forest biomass carbon pools in China and their significance , 2001 .

[16]  D. Lu,et al.  Change detection techniques , 2004 .

[17]  Richard Stone,et al.  Ecologists Report Huge Storm Losses in China's Forests , 2008, Science.

[18]  P. Leitão,et al.  Drivers of forest harvesting intensity patterns in Europe , 2014 .

[19]  J. Townshend,et al.  Impact of sensor's point spread function on land cover characterization: Assessment and deconvolution , 2002 .

[20]  Chris E. Jordan,et al.  Attribution of disturbance change agent from Landsat time-series in support of habitat monitoring in the Puget Sound region, USA , 2015 .

[21]  Chengquan Huang,et al.  Improving estimates of forest disturbance by combining observations from Landsat time series with U.S. Forest Service Forest Inventory and Analysis data , 2014 .

[22]  Stephen V. Stehman,et al.  Design and Analysis for Thematic Map Accuracy Assessment: Fundamental Principles , 1998 .

[23]  J. Vogelmann,et al.  Monitoring gradual ecosystem change using Landsat time series analyses: case studies in selected forest and rangeland ecosystems , 2012 .

[24]  Pol Coppin,et al.  Review ArticleDigital change detection methods in ecosystem monitoring: a review , 2004 .

[25]  C. Woodcock,et al.  Continuous monitoring of forest disturbance using all available Landsat imagery , 2012 .

[26]  Scott L. Powell,et al.  Forest Disturbance and North American Carbon Flux , 2008 .

[27]  Changzhu Li,et al.  The Great 2008 Chinese Ice Storm: Its Socioeconomic–Ecological Impact and Sustainability Lessons Learned , 2011 .

[28]  S. Goward,et al.  Dynamics of national forests assessed using the Landsat record: Case studies in eastern United States , 2009 .

[29]  Chengquan Huang,et al.  Assessing rates of forest change and fragmentation in Alabama, USA, using the vegetation change tracker model , 2009 .

[30]  D. Turner,et al.  Effects of harvest, fire, and pest/pathogen disturbances on the West Cascades ecoregion carbon balance , 2015, Carbon Balance and Management.

[31]  Bernard J. Lewis,et al.  Effects of logging intensity on structure and composition of a broadleaf-Korean pine mixed forest on Changbai Mountains, northeast China , 2016, Chinese Geographical Science.

[32]  Chengquan Huang,et al.  Quantifying Live Aboveground Biomass and Forest Disturbance of Mountainous Natural and Plantation Forests in Northern Guangdong, China, Based on Multi-Temporal Landsat, PALSAR and Field Plot Data , 2016, Remote. Sens..

[33]  Martha C. Anderson,et al.  Landsat-8: Science and Product Vision for Terrestrial Global Change Research , 2014 .

[34]  Douglas K. Bolton,et al.  Evaluating an Automated Approach for Monitoring Forest Disturbances in the Pacific Northwest from Logging, Fire and Insect Outbreaks with Landsat Time Series Data , 2014 .

[35]  S. Goward,et al.  An automated approach for reconstructing recent forest disturbance history using dense Landsat time series stacks , 2010 .

[36]  Chengquan Huang,et al.  Use of remote sensing coupled with a vegetation change tracker model to assess rates of forest change and fragmentation in Mississippi, USA , 2009 .