Evaluating natural experiments in ecology: using synthetic controls in assessments of remotely sensed land treatments.

Many important ecological phenomena occur on large spatial scales and/or are unplanned and thus do not easily fit within analytical frameworks that rely on randomization, replication, and interspersed a priori controls for statistical comparison. Analyses of such large-scale, natural experiments are common in the health and econometrics literature, where techniques have been developed to derive insight from large, noisy observational data sets. Here, we apply a technique from this literature, synthetic control, to assess landscape change with remote sensing data. The basic data requirements for synthetic control include (1) a discrete set of treated and untreated units, (2) a known date of treatment intervention, and (3) time series response data that include both pre- and post-treatment outcomes for all units. Synthetic control generates a response metric for treated units relative to a no-action alternative based on prior relationships between treated and unexposed groups. Using simulations and a case study involving a large-scale brush-clearing management event, we show how synthetic control can intuitively infer treatment effect sizes from satellite data, even in the presence of confounding noise from climate anomalies, long-term vegetation dynamics, or sensor errors. We find that accuracy depends on the number and quality of potential control units, highlighting the importance of selecting appropriate control populations. Although we consider the synthetic control approach in the context of natural experiments with remote sensing data, we expect the methodology to have wider utility in ecology, particularly for systems with large, complex, and poorly replicated experimental units.

Synergistic soil, land use, and climate influences on wind erosion on the Colorado Plateau: Implications for management.

Two decades of drought in the southwestern USA are spurring concerns about increases in wind erosion, dust emissions, and associated impacts on ecosystems, agriculture, human health, and water supply. Different avenues of investigation into primary drivers of wind erosion and dust have yielded mixed results depending on the spatial and temporal sensitivity of the evidence. We monitored passive aeolian sediment traps from 2017 to 2020 across eighty-one sites near Moab, Utah to understand patterns of sediment flux. At measurement sites, we collated climate, soil, topography and vegetation spatial layers to better understand the context of wind erosion and then combined these data with field observations of land use in models to characterize the influence of cattle grazing, oil and gas well pads, and vehicle/heavy equipment disturbance that potentially drive both exposure of bare soil and increases in erodible sediment supply that increase vulnerability to erosion. Disturbed areas with low soil calcium carbonate content yielded high sediment transport in dry years, but notably areas with little disturbance and low bare soil exposure had much less activity. Cattle grazing had the largest land use association with erosional activity with analyses suggesting that both herbivory and trampling from cattle could be drivers. The amount and distribution of bare soil exposure from new sub-annual fractional cover remote sensing products proved very helpful in mapping erosion, and new predictive maps informed by field data are presented to help depict spatial patterns of wind erosion activity. Our results suggest that despite the magnitude of current droughts, minimizing surface disturbance in vulnerable soils can mitigate a large portion of dust emissions. Results can help land managers identify eroding areas where disturbance reduction and soil surface protection measures can be prioritized.

Assessing vegetation recovery from energy development using a dynamic reference approach.

Ecologically relevant references are useful for evaluating ecosystem recovery, but references that are temporally static may be less useful when environmental conditions and disturbances are spatially and temporally heterogeneous. This challenge is particularly acute for ecosystems dominated by sagebrush (Artemisia spp.), where communities may require decades to recover from disturbance. We demonstrated application of a dynamic reference approach to studying sagebrush recovery using three decades of sagebrush cover estimates from remote sensing (1985-2018). We modelled recovery on former oil and gas well pads (n = 1200) across southwestern Wyoming, USA, relative to paired references identified by the Disturbance Automated Reference Toolset. We also used quantile regression to account for unmodelled heterogeneity in recovery, and projected recovery from similar disturbance across the landscape. Responses to weather and site-level factors often differed among quantiles, and sagebrush recovery on former well pads increased more when paired reference sites had greater sagebrush cover. Little (<5%) of the landscape was projected to recover within 100 years for low to mid quantiles, and recovery often occurred at higher elevations with cool and moist annual conditions. Conversely, 48%-78% of the landscape recovered quickly (within 25 years) for high quantiles of sagebrush cover. Our study demonstrates advantages of using dynamic reference sites when studying vegetation recovery, as well as how additional inferences obtained from quantile regression can inform management.

Disturbance automated reference toolset (DART): Assessing patterns in ecological recovery from energy development on the Colorado Plateau.

A new disturbance automated reference toolset (DART) was developed to monitor human land surface impacts using soil-type and ecological context. DART identifies reference areas with similar soils, topography, and geology; and compares the disturbance condition to the reference area condition using a quantile-based approach based on a satellite vegetation index. DART was able to represent 26-55% of variation of relative differences in bare ground and 26-41% of variation in total foliar cover when comparing sites with nearby ecological reference areas using the Soil Adjusted Total Vegetation Index (SATVI). Assessment of ecological recovery at oil and gas pads on the Colorado Plateau with DART revealed that more than half of well-pads were below the 25th percentile of reference areas. Machine learning trend analysis of poorly recovering well-pads (quantile<0.23) had out-of-bag error rates between 37 and 40% indicating moderate association with environmental and management variables hypothesized to influence recovery. Well-pads in grasslands (median quantile [MQ]=13%), blackbrush (Coleogyne ramosissima) shrublands (MQ=18%), arid canyon complexes (MQ=18%), warmer areas with more summer-dominated precipitation, and state administered areas (MQ=12%) had low recovery rates. Results showcase the usefulness of DART for assessing discrete surface land disturbances, and highlight the need for more targeted rehabilitation efforts at oil and gas well-pads in the arid southwest US.