Overview
Vegetation change has been observed across Arctic and boreal regions. Studies have often documented large-scale greening trends, but they have also identified areas of browning or shifts between greening and browning (see Box 1 for definitions of greening and browning) over varying spatial extents and time periods (e.g., Bhatt et al., 2013; Epstein et al., 2018; Lucht et al., 2002; Myneni et al., 1997; Nemani et al., 2003). At the same time, though, there are large portions of these ecosystems that have not exhibited measurable trends in greening or browning (e.g., Goetz et al., 2005). These findings have fueled many questions about the drivers of vegetation dynamics, how trends are measured, and potential implications of vegetation change at local to global scales.
In December 2018, the Polar Research Board, in collaboration with the Board on Life Sciences of the National Academies of Sciences, Engineering, and Medicine, convened a workshop to discuss opportunities to improve understanding of greening and browning trends and drivers and the implications of these vegetation changes. The discussions included a close look at many of the methodological approaches used to evaluate greening and browning, as well as exploration of newer technologies that may help advance the science.
Climate change is a driver of greening and browning trends at high latitudes and was discussed at the workshop. Annual air temperatures in this region are rising at more than twice the global average (Overland et al., 2018).1 This warming has been shown to contribute directly to greening trends in many areas of the tundra by stimulating plant growth and shifting the landscape toward increased plant (including shrub) cover (e.g., Epstein et al., 2018; Myers-Smith et al., 2019b). There are also strong linkages between warming and other known drivers of vegetation change, many of which manifest as disturbances that result in land cover change. Fire is a major disturbance in boreal forest that has increased in extent and severity with warming (Hanes et al., 2018), and there is increasing concern that fires will also become more frequent in the tundra. Browning trends in boreal forests have been linked to fire (Goetz et al., 2005; Ju and Masek, 2016), as well as to climate-related drought conditions (Beck et al., 2011). Insect infestations also cause browning as trees are either killed or defoliated. Recent work has shown an increase in spruce beetle outbreaks in Alaska that may be affected by warmer temperatures (Bentz et al., 2010), as well as a northward expansion of the range of the historically destructive mountain pine beetle in Canada, which suggests the potential for increased boreal forest browning in the future. Harvest of high-productivity tree species, particularly in Canada, was also discussed as a common disturbance that decreases forest cover and can contribute to browning trends.
The potential implications of vegetation change and related greening and browning patterns extend from local to global scales and can affect physical and ecological processes, climate feedbacks, and communities. The Earth’s surface energy balance is strongly affected by albedo (reflectance), with vegetation composition at high latitudes having a particularly strong impact. Snow-covered tundra landscapes and boreal forest areas where deciduous tree species (that are leafless during the winter) are dominant exhibit relatively high albedo because of reflectance off the snow (Loranty et al., 2014; Shuman et al., 2011) when the sun is shining. In contrast, areas with tall shrubs and evergreen conifer forest maintain a vegetated landscape above the snow, creating a much darker land surface resulting in lower albedo.
___________________
During the workshop, views were exchanged on patterns of greening and browning which can be associated with shifts in plant species composition, which in turn can lead to altered plant-soil-microbial feedbacks that influence long-term ecosystem structure and function. These shifts often occur following a disturbance and initially exhibit browning followed by greening as the ecosystem recovers either to a state similar to pre-disturbance, or to a new state. In boreal forest, a shift in states from a historically black spruce-dominated, low-resource environment often underlain by permafrost to a deciduous, more highly resource-dependent, warmer soil environment affects carbon storage, nutrient cycling, permafrost thaw, and greenhouse gas release, among other processes (Johnstone et al., 2010).
Vegetation change can also affect wildlife and hunting and have other societal impacts. Researchers are currently investigating linkages between migratory caribou, preferred food sources, and trends in greening and browning to discern whether vegetation change is contributing to a general decline in caribou populations in many areas. Vegetation change, particularly increasing shrub cover, was also discussed as a possible hindrance to hunters’ ability to visually find animals and as a negative impact on travel to hunting areas. A benefit of increasing shrub abundance around villages at high latitudes may be the increased use of shrubs as fuel, but at the same time, shrubs could hinder bifacial solar panel efficiency in the winter if shrubs reduce reflectance off the snow.
There are many commonly used methodologies applied to evaluate greening and browning trends and to identify drivers. These include remote sensing platforms (satellites, airborne campaigns, drones, etc.) and field-based approaches, which include measurements of individual plants or characterization of plot-level information such as growth, percentage cover, leaf greenup and senescence, as well as eddy covariance, which measures gas fluxes. Remote sensing is a useful tool for evaluating greening and browning trends at high latitudes because it can capture information across vast regions that are largely inaccessible over land. Many remote sensing platforms also provide relatively long records that can be
analyzed to determine trends. A tradeoff, however, is that these datasets are often at coarse spatial and temporal resolution and therefore may be unable to detect smaller-scale changes that are observable when using field-based methods. Field measurements allow for fine-resolution evaluation of productivity drivers and may identify changes that are only discernable at a local scale. Field-collected data are also important for validating findings of remote sensing studies.
The importance of the scale at which data are collected was a recurrent theme throughout the workshop. Participants emphasized that moving toward increased collection of data at finer spatial and temporal scales, with improved bridging of information across scales, would help to address observable inconsistencies that currently exist when comparing greening and browning patterns and trends across datasets. Expanded efforts to integrate new technologies and tools that have become available could also advance understanding.
Finally, utilization of a broader range of metrics to evaluate greening and browning that extend beyond those commonly used, in particular the remote sensing normalized difference vegetation index (NDVI), could allow for more detailed information that links greening and browning patterns to drivers and more direct comparisons of methodological approaches.
This page intentionally left blank.