Holocene history of landscape instability in Iceland: Can we deconvolve the impacts of climate, volcanism and human activity?

Biogeochemical proxy records from Icelandic lake sediment track large-scale shifts in North Atlantic Holocene climate and highlight the impact that North Atlantic Ocean- and atmospheric circulation has on Iceland's climate and environment. Following Early Holocene warmth, centennial-scale climate change is superimposed on millennial-scale cooling, culminating in the transition to the Little Ice Age (∼1300–1900 CE). Although the long-term cooling trend is presumably driven by variations in Earth's orbit and the concomitant decline in Northern Hemisphere (NH) summer insolation, the centennial-scale variability has been linked to the strength of the Atlantic Meridional Overturning Circulation (AMOC), volcanism coupled with sea ice/ocean related feedbacks, internal modes of atmospheric variability, and plausibly variations in solar irradiance. One manifestation of these regional climate changes on Iceland is the intensification of soil erosion, resulting in the degradation of ecosystems and landscape. In recent millennia, persistent and severe soil erosion has also been linked to human impact on the environment following the settlement ∼870 CE, rapid population growth, introduction of livestock and the poorly consolidated nature of tephra dominated soils. Lake proxy composite records suggest that although event-dominated landscape instability and soil erosion from the Early to Middle Holocene were likely triggered by large volcanic eruptions, the landscape was capable of recovering. However, a threshold was reached ∼5 ka BP, resulting in a state change whereby the Icelandic landscape could no longer fully recover from cold-events and/or tephra fall. Landscape sensitivity to climate further intensified at ∼1.5 ka BP as identified by regime shift analysis. Hence, widespread and irreversible soil erosion began several centuries before the acknowledged settlement of Iceland, with a second acceleration ∼1250 CE. A 2 ka fully coupled climate transient simulation using CESM1.1 shows a ∼0.5 °C reduction in summer temperature around Iceland in the first millennium CE, consistent with increased landscape instability and soil erosion in Iceland. A second phase of persistent summer cooling in the model occurs after 1150 CE, with stronger cooling after 1450 CE, reaching a maximum shortly after 1850 CE, ∼1 °C lower than at the start of the simulation. Our results suggest that natural variations in regional climate and volcanism are likely responsible for soil erosion prior to human impact, with intensification of these processes following settlement particularly during the cooling associated with the Little Ice Age. Given that the conclusions drawn in this review diverge from the standard paradigm of human-induced soil erosion history in Iceland, research should continue to focus on this complex question from multiple disciplines. In particular, a combination of emerging biogeochemical techniques (e.g. lipid biomarkers and ancient DNA) may be best poised to test and quantify the relative roles of natural environmental variables and human settlement in the history of soil erosion on Iceland.