Search

Invasive annual grasses, such as cheatgrass (Bromus tectorum L.), have proliferated in dryland ecosystems of the western United States, promoting increased fire activity and reduced biodiversity that can be detrimental to socio-environmental systems. Monitoring exotic annual grass cover and dynamics over large areas requires the use of remote sensing that can support early detection and rapid response initiatives. However, few studies have leveraged remote sensing technologies and computing frameworks capable of providing rangeland managers with maps of exotic annual grass cover at relatively high spatiotemporal resolutions and near real-time latencies. Here, we developed a system for automated mapping of invasive annual grass (%) cover using in situ observations, harmonized Landsat and Sentinel-2 (HLS) data, maps of biophysical variables, and machine learning techniques. A robust and automated cloud, cloud shadow, water, and snow/ice masking procedure (mean overall accuracy >81%) was implemented using time-series outlier detection and data mining techniques prior to spatiotemporal interpolation of HLS data via regression tree models (r = 0.94; mean absolute error (MAE) = 0.02). Weekly, cloud-free normalized difference vegetation index (NDVI) image composites (2016−2018) were used to construct a suite of spectral and phenological metrics (e.g., start and end of season dates), consistent with information derived from Moderate Resolution Image Spectroradiometer (MODIS) data. These metrics were incorporated into a data mining framework that accurately (r = 0.83; MAE = 11) modeled and mapped exotic annual grass (%) cover throughout dryland ecosystems in the western United States at a native, 30-m spatial resolution. Our results show that inclusion of weekly HLS time-series data and derived indicators improves our ability to map exotic annual grass cover, as compared to distribution models that use MODIS products or monthly, seasonal, or annual HLS composites as primary inputs. This research fills a critical gap in our ability to effectively assess, manage, and monitor drylands by providing a framework that allows for an accurate and timely depiction of land surface phenology and exotic annual grass cover at spatial and temporal resolutions that are meaningful to local resource managers.

Wetlands are critical terrestrial ecosystems in Alaska, covering ~177,000 km2 , an area greater than all the wetlands in the remainder of the United States. To assess the relative influence of changing climate, atmospheric carbon dioxide (CO2 ) concentration, and fire regime on carbon balance in wetland ecosystems of Alaska, a modeling framework that incorporates a fire disturbance model and two biogeochemical models was used. Spatially explicit simulations were conducted at 1-km resolution for the historical period (1950-2009) and future projection period (2010-2099). Simulations estimated that wetland ecosystems of Alaska lost 175 Tg carbon (C) in the historical period. Ecosystem C storage in 2009 was 5,556 Tg, with 89% of the C stored in soils. The estimated loss of C as CO2 and biogenic methane (CH4 ) emissions resulted in wetlands of Alaska increasing the greenhouse gas forcing of climate warming. Simulations for the projection period were conducted for six climate change scenarios constructed from two climate models forced under three CO2 emission scenarios. Ecosystem C storage averaged among climate scenarios increased 3.94 Tg C/yr by 2099, with variability among the simulations ranging from 2.02 to 4.42 Tg C/yr. These increases were driven primarily by increases in net primary production (NPP) that were greater than losses from increased decomposition and fire. The NPP increase was driven by CO2 fertilization (~5% per 100 parts per million by volume increase) and by increases in air temperature (~1% per °C increase). Increases in air temperature were estimated to be the primary cause for a projected 47.7% mean increase in biogenic CH4 emissions among the simulations (~15% per °C increase). Ecosystem CO2 sequestration offset the increase in CH4 emissions during the 21st century to decrease the greenhouse gas forcing of climate warming. However, beyond 2100, we expect that this forcing will ultimately increase as wetland ecosystems transition from being a sink to a source of atmospheric CO2 because of (1) decreasing sensitivity of NPP to increasing atmospheric CO2 , (2) increasing availability of soil C for decomposition as permafrost thaws, and (3) continued positive sensitivity of biogenic CH4 emissions to increases in soil temperature.

We summarize the results of a recent interagency assessment of land carbon dynamics in Alaska, in which carbon dynamics were estimated for all major terrestrial and aquatic ecosystems for the historical period (1950-2009) and a projection period (2010-2099). Between 1950 and 2009, upland and wetland (i.e., terrestrial) ecosystems of the state gained 0.4 Tg C/yr (0.1% of net primary production, NPP), resulting in a cumulative greenhouse gas radiative forcing of 1.68 × 10-3 W/m2 . The change in carbon storage is spatially variable with the region of the Northwest Boreal Landscape Conservation Cooperative (LCC) losing carbon because of fire disturbance. The combined carbon transport via various pathways through inland aquatic ecosystems of Alaska was estimated to be 41.3 Tg C/yr (17% of terrestrial NPP). During the projection period (2010-2099), carbon storage of terrestrial ecosystems of Alaska was projected to increase (22.5-70.0 Tg C/yr), primarily because of NPP increases of 10-30% associated with responses to rising atmospheric CO2 , increased nitrogen cycling, and longer growing seasons. Although carbon emissions to the atmosphere from wildfire and wetland CH4 were projected to increase for all of the climate projections, the increases in NPP more than compensated for those losses at the statewide level. Carbon dynamics of terrestrial ecosystems continue to warm the climate for four of the six future projections and cool the climate for only one of the projections. The attribution analyses we conducted indicated that the response of NPP in terrestrial ecosystems to rising atmospheric CO2 (~5% per 100 ppmv CO2 ) saturates as CO2 increases (between approximately +150 and +450 ppmv among projections). This response, along with the expectation that permafrost thaw would be much greater and release large quantities of permafrost carbon after 2100, suggests that projected carbon gains in terrestrial ecosystems of Alaska may not be sustained. From a national perspective, inclusion of all of Alaska in greenhouse gas inventory reports would ensure better accounting of the overall greenhouse gas balance of the nation and provide a foundation for considering mitigation activities in areas that are accessible enough to support substantive deployment.

High-latitude regions are experiencing rapid and extensive changes in ecosystem composition and function as the result of increases in average air temperature. Increasing air temperatures have led to widespread thawing and degradation of permafrost, which in turn has affected ecosystems, socioeconomics, and the carbon cycle of high latitudes. Here we overcome complex interactions among surface and subsurface conditions to map near-surface permafrost through decision and regression tree approaches that statistically and spatially extend field observations using remotely sensed imagery, climatic data, and thematic maps of a wide range of surface and subsurface biophysical characteristics. The data fusion approach generated medium-resolution (30-m pixels) maps of near-surface (within 1 m) permafrost, active-layer thickness, and associated uncertainty estimates throughout mainland Alaska. Our calibrated models (overall test accuracy of ~ 85%) were used to quantify changes in permafrost distribution under varying future climate scenarios assuming no other changes in biophysical factors. Models indicate that near-surface permafrost underlies 38% of mainland Alaska and that near-surface permafrost will disappear on 16 to 24% of the landscape by the end of the 21st Century. Simulations suggest that near-surface permafrost degradation is more probable in central regions of Alaska than more northerly regions. Taken together, these results have obvious implications for potential remobilization of frozen soil carbon pools under warmer temperatures. Additionally, warmer and drier conditions may increase fire activity and severity, which may exacerbate rates of permafrost thaw and carbon remobilization relative to climate alone. The mapping of permafrost distribution across Alaska is important for land-use planning, environmental assessments, and a wide-array of geophysical studies.

Quantification of aboveground biomass AGB in Alaska’s boreal forest is essential to the accurate evaluation of terrestrial carbon stocks and dynamics in northern high-latitude ecosystems. Our goal was to map AGB at 30 m resolution for the boreal forest in the Yukon River Basin of Alaska using Landsat data and ground measurements. We acquired Landsat images to generate a 3-year 2008–2010 composite of top-of-atmosphere reflectance for six bands as well as the brightness temperature BT. We constructed a multiple regression model using field-observed AGB and Landsat-derived reflectance, BT, and vegetation indices. A basin-wide boreal forest AGB map at 30 m resolution was generated by applying the regression model to the Landsat composite. The fivefold cross-validation with field measurements had a mean absolute error MAE of 25.7 Mg ha−1 relative MAE 47.5% and a mean bias error MBE of 4.3 Mg ha−1 relative MBE 7.9%. The boreal forest AGB product was compared with lidar-based vegetation height data; the comparison indicated that there was a significant correlation between the two data sets.

It is important to understand how upland ecosystems of Alaska, which are estimated to occupy 84% of the state (i.e., 1,237,774 km2 ), are influencing and will influence state-wide carbon (C) dynamics in the face of ongoing climate change. We coupled fire disturbance and biogeochemical models to assess the relative effects of changing atmospheric carbon dioxide (CO2 ), climate, logging and fire regimes on the historical and future C balance of upland ecosystems for the four main Landscape Conservation Cooperatives (LCCs) of Alaska. At the end of the historical period (1950-2009) of our analysis, we estimate that upland ecosystems of Alaska store ~50 Pg C (with ~90% of the C in soils), and gained 3.26 Tg C/yr. Three of the LCCs had gains in total ecosystem C storage, while the Northwest Boreal LCC lost C (-6.01 Tg C/yr) because of increases in fire activity. Carbon exports from logging affected only the North Pacific LCC and represented less than 1% of the state's net primary production (NPP). The analysis for the future time period (2010-2099) consisted of six simulations driven by climate outputs from two climate models for three emission scenarios. Across the climate scenarios, total ecosystem C storage increased between 19.5 and 66.3 Tg C/yr, which represents 3.4% to 11.7% increase in Alaska upland's storage. We conducted additional simulations to attribute these responses to environmental changes. This analysis showed that atmospheric CO2 fertilization was the main driver of ecosystem C balance. By comparing future simulations with constant and with increasing atmospheric CO2 , we estimated that the sensitivity of NPP was 4.8% per 100 ppmv, but NPP becomes less sensitive to CO2 increase throughout the 21st century. Overall, our analyses suggest that the decreasing CO2 sensitivity of NPP and the increasing sensitivity of heterotrophic respiration to air temperature, in addition to the increase in C loss from wildfires weakens the C sink from upland ecosystems of Alaska and will ultimately lead to a source of CO2 to the atmosphere beyond 2100. Therefore, we conclude that the increasing regional C sink we estimate for the 21st century will most likely be transitional.

Modern climate change in Alaska has resulted in widespread thawing of permafrost, increased fire activity, and extensive changes in vegetation characteristics that have significant consequences for socioecological systems. Despite observations of the heightened sensitivity of these systems to change, there has not been a comprehensive assessment of factors that drive ecosystem changes throughout Alaska. Here we present research that improves our understanding of the main drivers of the spatiotemporal patterns of carbon dynamics using in situ observations, remote sensing data, and an array of modeling techniques. In the last 60 yr, Alaska has seen a large increase in mean annual air temperature (1.7°C), with the greatest warming occurring over winter and spring. Warming trends are projected to continue throughout the 21st century and will likely result in landscape-level changes to ecosystem structure and function. Wetlands, mainly bogs and fens, which are currently estimated to cover 12.5% of the landscape, strongly influence exchange of methane between Alaska's ecosystems and the atmosphere and are expected to be affected by thawing permafrost and shifts in hydrology. Simulations suggest the current proportion of near-surface (within 1 m) and deep (within 5 m) permafrost extent will be reduced by 9-74% and 33-55% by the end of the 21st century, respectively. Since 2000, an average of 678 595 ha/yr was burned, more than twice the annual average during 1950-1999. The largest increase in fire activity is projected for the boreal forest, which could result in a reduction in late-successional spruce forest (8-44%) and an increase in early-successional deciduous forest (25-113%) that would mediate future fire activity and weaken permafrost stability in the region. Climate warming will also affect vegetation communities across arctic regions, where the coverage of deciduous forest could increase (223-620%), shrub tundra may increase (4-21%), and graminoid tundra might decrease (10-24%). This study sheds light on the sensitivity of Alaska's ecosystems to change that has the potential to significantly affect local and regional carbon balance, but more research is needed to improve estimates of land-surface and subsurface properties, and to better account for ecosystem dynamics affected by a myriad of biophysical factors and interactions.

Drylands are the habitat and source of livelihood for about two fifths of the world’s population and are highly susceptible to climate and anthropogenic change. To understand the vulnerability of drylands to changing environmental conditions, land managers need to effectively monitor rates of past change and remote sensing offers a cost-effective means to assess and manage these vast landscapes. Here, we present a novel approach to accurately monitor land-surface phenology in drylands of the Western United States using a regression tree modeling framework that combined information collected by the Operational Land Imager (OLI) onboard Landsat 8 and the Multispectral Instrument (MSI) onboard Sentinel-2. This highly-automatable approach allowed us to precisely characterize seasonal variations in spectral vegetation indices with substantial agreement between observed and predicted values (R2 = 0.98; Mean Absolute Error = 0.01). Derived phenology curves agreed with independent eMODIS phenological signatures of major land cover types (average r-value = 0.86), cheatgrass cover (average r-value = 0.96), and growing season proxies for vegetation productivity (R2 = 0.88), although a systematic bias towards earlier maturity and senescence indicates enhanced monitoring capabilities associated with the use of harmonized Landsat-8 Sentinel-2 data. Overall, our results demonstrate that observations made by the MSI and OLI can be used in conjunction to accurately characterize land-surface phenology and exclusion of imagery from either sensor drastically reduces our ability to monitor dryland environments. Given the declines in MODIS performance and forthcoming decommission with no equivalent replacement planned, data fusion approaches that integrate observations from multispectral sensors will be needed to effectively monitor dryland ecosystems. While the synthetic image stacks are expected to be locally useful, the technical approach can serve a wide variety of applications such as invasive species and drought monitoring, habitat mapping, production of phenology metrics, and land-cover change modeling.