Australia: The Land Where Time Began

A biography of the Australian continent 

Anomalous Arctic Warming Linked to Reduced North American Terrestrial Primary Productivity

Terrestrial productivity has been enhanced by warming temperatures in the Northern Hemisphere. Since 1990 these events have been linked to anomalous warming in the Arctic, and Kim et al. suggest they may affect Terrestrial processes. In this paper Kim et al. analyse multiple observation data sets and numerical model simulations to evaluate links between temperatures in the Arctic and primary productivity. The study found that in the Arctic positive temperature anomalies in the spring have resulted in negative anomalies in the gross primary productivity across most of North America over the last 3 decades, which is a net decline of primary productivity of 0.31 PgC/year (petagrams of carbon) across the continent. There are 2 factors that mainly explain this decline: conditions of severe cold in northern North America and lower precipitation in the South Central United States. Also, it is revealed by crop-yield data for the US that yields declined by an average of approximately 1-4 %, and individual states experienced declines of up to 20 %, in years of anomalous warming in the Arctic. It was concluded by Kim et al. that strengthening of the warming anomalies in the Arctic over past decades has reduced productivity over North America remotely.

Climate changes over the last few decades that are a result of anthropogenic forcings and processes of natural feedback have affected the productivity of terrestrial ecosystems around the Earth.  (Myneni et al., 1997; Zhu et al., 2016). An increase in terrestrial gross primary productivity  (GPP), such as the expansion of shrub cover, enhanced photosynthesis by vegetation and the lengthening of the growing season, especially at high latitudes, is one of the main consequences of terrestrial ecosystem changes (Bhatt et al.,. 2010; Hinzman et al., 2005). These positive changes in vegetation productivity at high latitudes are related closely to the recent warming across the high latitude regions (Piao et al., 2008). The Arctic has recently shown a remarkably rapid temperature trend compared with other regions; as indicated by the observed records; that is known as the Arctic Amplification (Kug et al., 2016; Francis & Vavrus, 2012; Cohen et al., 2014; Screen & Simonds, 2016; Wallace et al., 2014). It has recently been reported, however, that anomalous warming of the Arctic can result in severe cold events at mid-latitudes (Kug et al., 2016; Kim et al., 2014). It is possible that mid-latitude terrestrial ecosystems are affected in the opposite direction via teleconnections that are induced by anomalous warming in the Arctic, even though warming of the Arctic regions has resulted in positive changes in high latitude productivity of vegetation. Understanding of the remote impacts of anomalous Arctic warming on terrestrial GPP at mid-latitudes is, however, not sufficient.

Atmospheric teleconnections linked to warming of the Arctic

It has recently been reported that regional temperatures anomalies in the Arctic are critical to an explanation of climate variations in downstream regions; such as the close relationship between the cold winters over North America via the development of downstream teleconnection and the Arctic anomalies over the East Siberian-Chukchi Sea (Kug et al., 2015; Wallace et al., 2014). In this paper Kim et al. used the Arctic temperature (ART) index, which was introduced in a previous study (Kug et al., 2015),  to represent the regional Arctic temperature anomalies that averaged temperature anomalies over the East Siberian-Chukchi Sea (160^oE -160^oW, 65^o-80^oN). The March ART index particularly shows the most significant relation with simultaneous and lagged temperature anomalies over North America; therefore this index is used to evaluate remote impacts on mid-latitude atmospheric conditions and terrestrial GPP over North America by regional Arctic temperature anomalies.

The first step was Examination of the atmospheric teleconnection pattern that is related to temperature anomalies in the Arctic, which is represented by the regressed circulation pattern with respect to the ART index for the period 1979-2015. Positive sea level pressure anomalies over Alaska might be a direct response to positive temperature anomalies in the Arctic. This anticyclonic flow that is induced by local forcing expands to the east due to strong low-level cloud cold advection. A distinctive anticyclone in the upper level that is located over Alaska shows an equivalent barotropic structure. Also, there are cyclonic and anticyclonic anomalies that are located in the downstream regions; this can be explained by the propagation of Rossby waves (Honda, Inoue & Yamane, 2009). As a result of the atmospheric teleconnection that is induced by anomalous Arctic warming, low-level anticyclone and upper level cyclone anomalies, by which favourable conditions for severe cold weather are provided, are deployed over North America.

Considerable surface anomalies are observed in the northern part of North America, while in in Alaska and East Siberia positive temperature anomalies are observed, which is consistent with the large-scale atmosphere pattern. An anomalous southwesterly wind is also observed along the east coast of the United States related to the anomalous anticyclone over the subtropical Atlantic, in contrast to the northerly wind in northern North America. An eastwards shift of the Great Plains low-level jet is indicated by the anomalous southwesterlies, which can lead to a dipole precipitation pattern by alteration of the transport of moisture (Higgins, Mo & Yao, 1998).

Terrestrial productivity anomalies linked to warming in the Arctic

Significant anomalies of temperature and precipitation have been observed over North America that are associated with variation in Arctic temperature. It is expected that anomalous temperature and precipitation changes over North America on a continental scale could affect terrestrial ecosystems. The relations of multiple data sets that are used as a proxy for terrestrial GPP with the ART index are analysed to examine the impact Arctic warming is having on terrestrial GPP over North America. It has been found that negative vegetation activity and terrestrial GPP are captured across North America that are related to anomalous Arctic warming. Significant changes in terrestrial GPP have been exhibited across an extensive area of North America, from Canada, in the broad coniferous forests, to Mexico, in the subtropical steppe. It is clear that in terms of spatial pattern among various data sets, that there is consistent vegetation activity, such as satellite remote sensing of the normalised difference vegetation index (NDVI) for the period 1982-2013, and the GPP that is the flux tower that is data-driven that is based on a model tree ensemble (MTE) for the period 1982-2011. Consistent results are shown for the terrestrial ecosystem models; i.e., the simulated multi-model ensemble (MME) GPP correlates negatively to the ART index. Kim et al. found that the terrestrial productivity anomalies that were induced by anomalous Arctic warming are not sensitive to the data period, though the trends of NDVI and GPP can be dependent on the data period. Also, consistent results were shown for the Earth system models, which took part in the Coupled Model Intercomparison Project Phase 5 (CMIP5). Consistent results were shown by the observed GPP variations from individual flux towers with large scale data in terms of negative GPP anomalies in the case of anomalous Arctic warming, in spite of limited observed sample size. It is indicated by both the data-driven and the process-driven GPP that a change of about -0.31PgC/year over North America (125^o-85^oW, 30^o-60^oN), even though interannual variability tends to be underestimated (Jung et al., 2011).

Kim et al. suggest that a major driver for the negative GPP would be the cold surface over North America that was associated with anomalous Arctic warming. Terrestrial GPP anomalies in the norther part of North America are related closely to anomalous Arctic warming temperature anomalies because temperate and boreal regions are composed of ecosystems that are limited by temperature. This is consistent with a previous study, which demonstrated that the weakening positive trends in vegetation spring and summer greenness, is related to temperature variations during spring in that region (Wang et al., 2011). In detail, the maximum NDI and GPP appear in the Great Lakes Basin in the northwest of the United States, while the maximum temperature anomaly is located in the Manitoba and Saskatchewan provinces of Canada to the northwest of the Great Lakes Basin. Kim et al. suggest that this may be attributed to the sensitivity difference of the GPP to the cold damage, the cold tolerance, of the vegetation, depending on the plant functional types (Kim et al., 2014). E.g., there is a high fraction of forest that is comprised of needleleaf, evergreen forest, which has cold tolerance that is relatively stronger than other classes of land cover. However, deciduous broadleaf forest and mixed forest, in which the sensitivity to cold damage is higher than in evergreen needleleaf forest are distributed mainly in the Great Lakes Basin (Kim et al., 2014); thereby, NDVI and GPP anomalies that are related to anomalous Arctic warming have shown a southeastwards shift pattern compared to temperature anomalies.

Terrestrial variation in the South Central United States might be related to precipitation anomalies with respect to anomalous Arctic warming, as well as the temperature effect. The reduction of the GPP in the southern part of North America is mainly accompanied by a decrease in precipitation in that region, as a result of water-limited ecosystems in that region (Nemani et al., 2013). The increased precipitation on the east coast of the United States does not contribute to increased terrestrial GPP, as the ecosystems in that region are not water limited, possibly as a result of enough climatological precipitation, which contrasts to the decreased precipitation in the South Central United States.

Also, monthly GPP anomalies to March Arctic temperature anomalies show that the impacts on the terrestrial GPP are at maximum in May and even maintained until early summer, in both flux tower data-driven and simulated results. Ecologically, biological stresses may result from environmental disturbances, such as plant cellular dehydration, low stomatal conductance, canopy development suppression, and leaf area, especially for early spring, as leaves that are newly emerged in spring are sensitive to cold (Hufkens et al., 2012) and drought (Noormets et al., 2008) stresses as a result of structural rigour, which is necessary to avoid cellular damage (Menzel & Fabian, 1999). This is consistent with previous studies in which it was argued that changes in the productivity of spring vegetation tend to affect the productivity in terrestrial systems in the succeeding months (Jeong et al.,. 2012; Kim et al., 2014). Also, significant anomalies are shown to the local temperature and precipitation in spring over some parts of North America, which suggests there is a role for atmospheric conditions in spring in interannual variability of terrestrial productivity in North America. Therefore, abiotic stresses, that include cold and drought stresses, during spring that result from anomalous Arctic warming simultaneously suppress productivity and even contribute to reduced annual productivity, possibly by lasting effects on the function of ecosystems.

Impacts on the US crop yield

It is revealed by the US-national-level crop yield data that annual yields of corn, soybeans and wheat declined by about 1.74, 3.96 and 3.62 %, respectively, in years of Arctic warming compared to years of Arctic cooling, which is consistent with cumulative annual terrestrial GPP. It is suggested by state-level differences of crop yield between Arctic warming and Arctic cooling cases that all 3 major crops mainly exhibit changes that are  negative in response to anomalous Arctic warming, albeit with some differences in crop fraction, sensitivity of crop to climatic conditions, and scheme of management between states (Butler & Huybers, 2013). The regions where a significant degree of crop yield reduction is displayed are consistent with the overall negative terrestrial productivity anomalies in North America which were obtained by remote sensing of NDVI, as well as by data-driven and process-driven GPP, which also includes areas of irrigated cropland in the southern part of the US. Crop yields are concurrently apparent in the Great Plains for all 3 crops, such as North Dakota (soybeans -0.24 and wheat -0.44 t/ha/yr.), South Dakota, Nebraska and Kansas (soybean -0.39t/ha/yr.).  Corn in the southern US displays the largest decrease in crop yield, especially in Texas (-1.11 t/ha/yr; which is about 20 % of the productivity in a normal year in this region). Kim et al. suggest the crop yield reductions in the southern US could be related to precipitation decreases), as is shown in the response of the atmosphere to anomalous arctic warming. This may be a result of crop productivity in dry areas of the southern US being crucially dependent on water resources. The decreasing crop yield in the northern Great Plains might be related to decreased temperature in spring, which is in contrast to the situation in the southern Great Plains. A few states in the northwestern US exhibit positive relations, especially for the yield of wheat, though crop yield changes related to anomalous Arctic warming show negative relations in most of the regions. This pattern can be explained by increases in precipitation in the northwestern US, which Kim et al. suggest may have a positive impact on the productivity of wheat (Tack, Barkley & Nalley, 2015).

Overall, this study has demonstrated for the first time an apparent link between temperature variations in the Arctic and mid-latitude agricultural productivity. As the understanding of large-scale circulation patterns can be useful for the improvement of the predictability of terrestrial productivity and crop yields (Cane et al., 1994; Kim et al., 2016; Hallett et al., 2004; Ciais et al., 2005; Bastos et al., 2016), the results of this study suggest the Arctic information could be used in the forecasting of agricultural productivity and a reduction of the uncertainty. In particular, as the variation of the interannual Arctic temperature has been distinctly stronger in recent decades as the Arctic sea ice (Kug et al., 2015; Kim et al., 2014) declines rapidly, it is suggested this variation in Arctic temperature may negatively impact human life in the form of adverse weather conditions as well as agricultural productivity over North America. Moreover, the simulated MME net ecosystem exchange anomaly that is associated with anomalous Arctic warming is about -0.1 PgC/yr, which is about 20 % of the interannual standard deviation range in the carbon sink of North America (King et al., 2015).

According to Kim et al. the current climate models tend to simulate the negative terrestrial GPP anomalies that are associated with the anomalous Arctic warming over North America reasonably well, though the detailed spatial pattern differs from the patterns that have been observed. Moreover, these negative GPP anomalies are also seen in future climate simulations, which suggest the relationship is robust. Of interest, the Arctic-related GPP anomalies are even stronger under future climatic conditions, especially in northwest North America. Kim et al suggest this is related to sensitivity of GPP anomalies to local temperature becomes stronger under greenhouse warming, and it is consistent with previous studies that have argued enhanced phenological frost damage in a warming climate (Gu et al., 2008; Rigby & Porporato; Augspurger, 2009; Augspurger, 2013). I.e., Arctic-induced cold stress in a warmer climate will damage more severely the ecosystem in that region. This result has delivered important implications for climate adaptation, though to obtain a general conclusion further investigation is required.

Sources & Further reading

1. Kim, J.-S., et al. (2017). "Reduced North American terrestrial primary productivity linked to anomalous Arctic warming." Nature Geosci 10(8): 572-576.