FINAL REPORT: ASSESSING POST-FIRE RECOVERY OF SAGEBRUSH-STEPPE RANGELANDS IN SOUTHEASTERN IDAHO (NNX08AO90G) Keith T. Weber and Kerynn Davis, editors Contributing Investigators Keith T. Weber, Principal Investigator ([email protected]), Idaho State University, GIS Training and Research Center, 921 S. 8th Ave., stop 8104, Pocatello, ID 83209-8104. Project web-site: http://giscenter.isu.edu/research/Techpg/nasa_postfire/template.htm All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means without the prior permission of the editor. ACKNOWLEGDEMENTS This study was made possible by a grant from the National Aeronautics and Space Administration Goddard Space Flight Center (NNX08AO90G). ISU would like to acknowledge the Idaho Delegation for their assistance in obtaining this grant. Recommended citation style: Davis, K. and K. T. Weber, 2011. 2009 Rangeland Vegetation Assessment at the O’Neal Ecological Reserve, Idaho. Pages 3-9 in K. T. Weber and K. Davis (Eds.), Final Report: Assessing Post-Fire Recovery of Sagebrush-Steppe Rangelands in Southeastern Idaho (NNX08AO90G). 252pp. Table of Contents Chapter Title (author) Page Executive Summary 1 1 2009 Rangeland Vegetation Assessment at the O’Neal Ecological Reserve, Idaho 3 (Davis and Weber) 2 2009 Range Vegetation Assessment in the Big Desert, Upper Snake River Plain, 11 Idaho (Studley and Weber) 3 2010 Rangeland Vegetation Assessment at the O’Neal Ecological Reserve, Idaho 21 (Davis et al.) 4 2010 Field Spectrometry Collection of Sagebrush at the O’Neal Ecological Reserve, 29 Idaho (Hanson and Weber) 5 Intercalibration and Evaluation of ResourceSat-1 and Landsat-5 NDVI (Anderson et 37 al.) 6 Assessing the Susceptibility of Semiarid Rangelands to Wildfires using Terra 47 MODIS and Landsat Thematic Mapper Data (Chen et al.) 7 Comparison of MODIS fPAR Products with Landsat-5 TM Derived fPAR over 69 Semiarid Rangelands of Idaho (Chen et al.) 8 Herbaceous Biomass Estimation from SPOT-5 Imagery in Semiarid Rangelands of 89 Idaho (Chen et al.) 9 NDVI Changes over a Calendar Year in the Rangelands of Southeast Idaho (Tedrow 105 and Weber) 10 Diurnal NDVI Fluctuations in Semiarid Rangelands (Weber and Chen) 117 11 Detection Thresholds for Rare, Spectrally Unique Targets within Semiarid 129 Rangelands (Weber and Chen) 12 Comparison of Atmospheric Correction Algorithms for Multispectral Satellite 143 Imagery (Weber) 13 Comparing Two Ground-Cover Measurement Methodologies for Semiarid 149 Rangelands (Weber et al.) 14 Evaluating Land Degradation Indicators in Semiarid Ecosystems Relative to Wildfire 161 (Weber and Chen) 15 Comparison of Image Resampling Techniques for Satellite Imagery (Studley and 185 Weber) 16 Applying Indigenous Pastoralist Experiences to Western Range Management 197 (Weber and Horst) 17 Detecting Dead Shrub Patches Using Remote Sensing Techniques in Southeast 211 Idaho (Hanson and Weber) 18 Quantifying Habitat Fragmentation in the Big Desert, Idaho (Hanson and Weber) 235 Final Report: Assessing Post-Fire Recovery of Sagebrush-Steppe Rangelands in Southeastern Idaho ASSESSING POST-FIRE RECOVERY OF SAGEBRUSH STEPPE RANGELANDS IN SOUTHEASTERN IDAHO Executive Summary Significant Findings and Achievements • Two primary objectives were addressed in this study, 1) determine the ability of geospatial technologies to reliably capture and characterize changes in the vegetation community following wildfire and 2) determine if the Big Desert study area in southeast Idaho is moving toward a state of desertification as a consequence of the 2006 Crystal fire disturbance. • Characterizing changes in the vegetation community following a wildfire disturbance was examined using MODIS, Landsat, and SPOT satellite imagery. In each case specific abilities of each sensor were revealed including the ability of MODIS fPAR data to reliably describe changes in primary productivity within semiarid savanna ecosystems (cf. chapters 6-8). • Semiarid savanna ecosystems are paradoxical. At first glance the untrained eye may be led to believe the sagebrush-steppe is a stagnant sea of unchanging shrubs. In reality however, these ecosystems exhibit tremendous changes throughout each year and each growing season (cf. chapter 9). Furthermore, the C3 plants found throughout the study area exhibit distinct diurnal patterns of respiration and photosynthesis (cf. chapter 10). A consequence of these research findings have direct implication for primary productivity modeling and suggest current primary productivity values may underestimate actual productivity of semiarid savanna ecosystems. • Assessing changes in land cover over extended temporal periods requires the use of a correspondingly lengthy dataset. Making valid comparisons across time further requires the use of complimentary datasets. Based upon these statements and suppositions it becomes clear that the Landsat data archive is arguably the most important dataset available to geospatial scientists. However, current Landsat satellites are experiencing difficulties and it is uncertain whether Landsat imagery will remain continuous through the successful launch of Landsat 8 in December of 2012. • An investigation into the complementary nature of Landsat and ResourceSat --a recommended interim alternative platform-- was conducted as part of this study. Results suggest with proper site-specific intercalibration, ResourceSat could serve adequately as an interim replacement for Landsat (cf. chapter 5). • The development of a long-term Landsat dataset (1984-2010) was initiated as part of this project. Using data from 2000-2009 an evaluation of various primary productivity indicators was conducted to assess the status of sagebrush-steppe rangelands prior to and following the 250,000 acre Crystal fire of 2006. Specific indicators used in this study include composite-NDVI , rain- use efficiency, water-use efficiency, and local net primary productivity scaling. Results suggest that while primary productivity is estimated to be relatively low across the Big Desert, it is also fairly stable when viewed from a long-term, decadal perspective. Results also underscore the importance of using long-term data (10 years longer) for this and other assessments as short-term (2-4 years) snapshots of change could be misleading (cf. chapter 14). 1 Final Report: Assessing Post-Fire Recovery of Sagebrush-Steppe Rangelands in Southeastern Idaho • The detection of landscape or regional land cover change is ultimately tied to fine-scale changes in bare ground, plant community species/structure, and current phenological-status at the time of data capture. Issues related to phenology can be effectively resolved through the use of composite-NDVI, but properly scaling data from plant scale (fine-scale) to landscape scale is very difficult. To address scaling issues we examined the ability of remote sensing technologies to detect various fine-scale landscape features such as small patches of dead shrubs (cf. chapters 4 and 17) and randomly located, rare and spectrally unique targets (cf. chapter 11). Results demonstrate various limitations of current remote sensing capabilities due to mixed pixels and the signal-to-noise ratio of the sensor. • Eight of the chapters included in this final report have been published (five) or are in review for publication (three) in a peer-reviewed professional journal. • Five graduate students were supported in whole or in part through this grant (Bhushan Gokhale [cf. chapters 5-8, and 13], Darci Hanson [cf. chapters 3, 4, 17, and 18], Mansoor Raza [cf. chapter 13], Heather Studley [cf. chapters 2 and 15]), and Linda Tedrow [cf. chapter 9]) as well as one undergraduate student (Kerynn Davis [cf. chapters 1 and 3]). Each of these students have completed their studies at Idaho State University and have graduated. 2 Final Report: Assessing Post-Fire Recovery of Sagebrush-Steppe Rangelands in Southeastern Idaho 2009 Rangeland Vegetation Assessment at the O’Neal Ecological Reserve, Idaho Kerynn Davis, Idaho State University. GIS Training and Research Center, 921 S. 8th Ave., Stop 8104, Pocatello, ID 83209-8104 Keith T. Weber, GISP. GIS Director, Idaho State University. GIS Training and Research Center, 921 S. 8th Ave., Stop 8104, Pocatello, ID 83209-8104. [email protected] ABSTRACT Vegetation data were collected at 30 randomly located sample points during June 2009. Data were collected using both ocular estimation and line-point intercept transects each describing fuel load and percent cover of grasses, forbs, shrubs, litter, microbial crust, bare ground, and weeds respectively. In the SHPG (Simulated Holistic Planned Grazing) grazing treatment of the O’Neal, percent cover of grass (2009=24.66%, 2008=13.84%), forbs (7%, 5.82%), and shrubs (12.33%, 11.1%) increased from 2008. The Rest-Rotation grazing allotment also saw increased percentage cover in grasses (22.66%, 8.96%), forbs (9.16%, 6.34%) and shrubs (13.83%, 11.26%). In the Total Rest grazing allotment, percent cover increased in grasses (28.33%, 12.27%), forbs (10.33%, 4.1%), and weeds (13.6%, 12.33%). Much of the changes observed are likely attributable to the increase in precipitation in 2009 (106.8 mm) relative to 2008 (9.2 mm). KEYWORDS: Vegetation, sampling, GIS, remote sensing, GPS, grazing treatment, land management 3 Final Report: Assessing Post-Fire Recovery of Sagebrush-Steppe Rangelands in Southeastern Idaho INTRODUCTION There are many factors that influence land cover changes. Wildfire has been, and will always