Quantifying drought-induced tree mortality in the open canopy woodlands of central Texas

Published Date

August 2016, Vol.181:54–64, doi:10.1016/j.rse.2016.03.027

Title 

Quantifying drought-induced tree mortality in the open canopy woodlands of central Texas

• Author 

• Amanda M. Schwantes ^a,,
• Jennifer J. Swenson ^a,
• Robert B. Jackson ^a,b,

• aNicholas School of the Environment, Duke University, Box 90328, Durham, NC 27708, USA
• bDepartment of Earth System Science, Woods Institute for the Environment, and Precourt Institute for Energy, Stanford University, Y2E2 Building, 379B, Stanford, CA 94305, USA

Received 7 October 2015. Revised 19 February 2016. Accepted 23 March 2016. Available online 16 April 2016.

Highlights

• •
  We quantify continuous drought-driven canopy loss across an area > 30,000 km².
• •
  We conduct a time-series analysis to identify areas with persistent canopy loss.
• •
  Orthophotos are used to train Landsat & validate with spatially independent data.
• •
  For scaling canopy loss we compare random forest and zero-inflated beta regression.
• •
  We scale field measurements to fine-scale (1-m) maps to regional (60-m) estimates.

Abstract

During 2011, Texas experienced a severe drought, which caused substantial tree mortality. Drought-induced tree mortality can have significant ecological impacts and is expected to increase in many locations with climate change. This disturbance is unique because it often is limited to only subtle and diffuse changes in forest cover. Thus we developed new methods to quantify drought-driven canopy loss using remotely sensed imagery, across a Landsat scene in central Texas (> 30,000 km²). First, fine-scale canopy loss maps were created by classifying 17 1-m orthophotos (each ~ 50 km²) from the US National Agriculture Imagery Program. These classifications were highly correlated (R² = 0.90) with field estimates of canopy cover loss measured in 21 plots at 4 sites across central Texas. These fine-scale canopy loss maps were then used to calibrate and validate coarser-scale Landsat imagery. In scaling up to create regional canopy loss maps, we assembled a Landsat time-series and separated mortality pixels experiencing persistent canopy loss from pixels with only background noise by applying the Landtrendr algorithm. We then estimated percent tree canopy loss within each of these mortality pixels by comparing two models capable of handling zero-inflated continuous proportions: random forest and a zero-or-one inflated beta (ZOIB) regression model. We found that the ZOIB regression model had the highest accuracy in predicting canopy loss (mean absolute error = 5.16%, root mean square error = 8.01%). The 2011 drought caused a decrease in canopy cover within the study area, equivalent to 1124 km² of canopy loss, ~ 10% of the 10,850 km² area of live canopy present before the drought. Our methods address the need to detect drought-induced tree mortality as extreme droughts are predicted to increase with climate change. More detailed canopy loss maps could then be used (1) to quantify potential impacts to carbon cycling, biophysics, and species compositions and (2) to understand the factors controlling tree mortality, now and in the future.

Graphical abstract

Keywords

• Drought
• Tree mortality
• Canopy loss
• Disturbance
• Die-back
• Change detection
• Zero-or-one inflated beta regression models
• Random forest
• Landsat time-series
• Landtrendr

---

Fig. 1.

Fig. 2.

 Table 1

Table 1.

 Table 2

Table 2.

Fig. 3.

 Table 3

Table 3.

 Table 4

Table 4.

 Table 5

Table 5.

Fig. 4.

Fig. 5.

• DOCX (81K)

• ⁎ 
  Corresponding author.

For further details log on website :
http://www.sciencedirect.com/science/article/pii/S0034425716301213