Validating a Landsat Time-Series of Fractional Component Cover Across Western U.S. Rangelands

Abstract

Western U.S. rangelands have been quantified as six fractional cover (0%-100%) components over the Landsat archive at a 30 m resolution, termed the "Back-in-Time" (BIT) dataset. Robust validation through space and time is needed to quantify product accuracy. Here, we used field data collected concurrently with high-resolution satellite (HRS) images over multiple locations (n = 42) and years. Field observations were used to train regression tree models, predicting the component cover across each HRS image. Our objectives were to evaluate the spatial and temporal relationships between HRS and BIT component cover and compare spatio-temporal climate responses. First, for each HRS site-year (n = 77) we averaged both the HRS and BIT predictions within each site separately and regressed the averages to quantify the temporal accuracy. Next, we regressed individual pixel values of corresponding HRS and BIT predictions to quantify the spatio-temporal accuracy. Results showed strong temporal correlations with an average R 2 of 0.63 and Root Mean Square Error (RMSE) of 5.47% as well as strong spatio-temporal correlations with an average R 2 of 0.52 and RMSE of 7.89% across components. Our approach increased the validation sample size relative to direct comparison of field observations. Validation results showed robust spatio-temporal relationships between HRS and BIT data, providing increased user confidence in the data. disturbance in the Northern Great Basin [4,5]. While most change in fractional component cover was small (<10% cover change over the study period), a large majority of pixels indicated at least some change [5]. The BIT suite was designed to capture spatially discrete abrupt change and pervasive gradual change [5] that is often ignored by the remote sensing community [6].Validation of any remotely sensed mapping application is critical to increasing user confidence in products, foster usage in management decisions, and determine the more robust estimate in cases of competing datasets [7,8]. However, validation is often challenging [3,6,[9][10][11], especially with time-series maps [12]. Validation is often somewhat subjective, relying on manual image interpretation on reference blocks [13], Google Earth time-series imagery [14,15], or case studies. Most validation methods are designed for thematic classes (e.g., [10,15]) that typically employ confusion matrices to understand accuracies [14]. An introduction of more mapping classes often results in weaker validation results [15]. By extension, validation of fractional component time-series pose the most difficult scenario, particularly in areas of subtle change. This is especially true in dryland ecosystems with frequently sparse vegetation canopies that increase the influence of soils and senesced vegetation and where only a scarce ground-based data network exists [16]. Major challenges to time-series validation include (1) validation datasets that are not directly comparable to the remotely sensed data, (2) sample size, spatial extent, or temporal extent of...

Projected change in rangeland fractional component cover across the sagebrush biome under climate change through 2085

“…The BIT data have been robustly validated by Rigge et al (2019b) who compared BIT data to a series of predictions made on high-resolution satellite imagery finding strong temporal relationships with a mean coefficient of determinations (R 2 ) of 0.63 and root-mean-square error (RMSE) of 5.47% and strong spatiotemporal correlations with a mean R 2 of 0.52 and RMSE of 7.89% across components. Data collected at longterm monitoring plots (n = 126) over a 10-yr period of 2006-2018 in southwestern Wyoming were compared to BIT data, again showing robust temporal and spatiotemporal correlations, with a mean R 2 of 0.46 across components (Shi et al 2020).…”

Application of Empirical Land-Cover Changes to Construct Climate Change Scenarios in Federally Managed Lands

“…Pixels were flagged as changed if thresholds were exceeded in both summer and fall images in a given year. Training data from the circa 2014 base were selected within the unchanged pixels in each year [28,34,35]. The training points were pooled across time to develop Cubist regression tree (RuleQuest Research 2008) models which used two seasons (summer and fall) of Landsat imagery, topographic data, and spectral indices as inputs [28,34].…”

“…A series of post-processing models were applied to ensure accurate post-burn trajectories and to eliminate noise and illogical change. For more details on the process, see Rigge et al [35].…”

Assessing vegetation recovery from energy development using a dynamic reference approach

“…First, we excluded well pads with <100% overlap with sagebrush remote sensing data. Second, we excluded well pads if mean sagebrush cover among reference pixels was ≤5.9% in any year following apparent reclamation, corresponding with the root mean square error for sagebrush estimates when compared with independent high resolution data (Rigge et al, 2019) and therefore potentially lacking sagebrush. Third, we used a LANDFIRE dataset for Existing Vegetation Type (LF 2.0.0; Rollins, 2009) with a crosswalk to Society of American Foresters‐Society for Range Management cover types to retain pads if at least one reference pixel was classified as “Mountain Big Sagebrush,” “Wyoming Big Sagebrush,” “Sagebrush‐Grass,” or “Big Sagebrush‐Bluebunch Wheatgrass.”…”