Keywords

8.1 The Need for Early Warning

While climate change will increase temperatures, desertification and frequency of drought across different parts of the world (IPCC 2007), and thus possibly increase the potential for dust storms (Wilcox 2012; Shao et al. 2011), the actual impact of climate change on dust storm frequency and severity is unknown. Impacts of climate change can be exacerbated by human activities which already contribute to the intensity and frequency of dust storms (Yang et al. 2007; Ginoux et al. 2010). Additionally, different regions of the globe have high interannual, as well as annual and decadal, variability of dust events, thus furthering the need for more research to be conducted over longer periods of time to analyse trends of occurrences and associated severity (Ganor et al. 2010; Goudie 2009). The origin of dust storms, whether natural or human, and how aerosol circulation patterns are affected, need to also be evaluated to understand the ultimate impact on the global climate (Ginoux et al. 2010). With this accumulation of information, more accurate forecasts of dust storm movements can be developed, the appropriate efforts to mitigate damage can be put into place and effective early warning can be communicated.

The dust cycle is a dynamic process of emission, transport, transformation, deposition and stabilisation that occurs at both local and global scales on varying time scales (Shao et al. 2011) (Fig. 8.1). A thin crust, formed by desert soils, most prevalent in areas between plants, helps to stablise the ground surface and create a natural resistance to wind erosion. The plants protect surrounding soils from wind and trap suspended soil particles (Urban et al. 2009; Steenburgh et al. 2012; Wilcox 2012). A disturbance to the crust or a reduction in vegetation cover increases the risk of a dust storm occurrence, as the loosened sediments are free to be picked up by high winds. Dust storms can result in the deposition of foreign sediments on crops, blocking sunlight and causing them to suffocate. Dust particles and the pollutants they carry can contribute to compromised air quality and human health and can also influence climatic patterns and the energy balance of the earth system (Shao et al. 2011).

Fig. 8.1
figure 1

Dust storm life cycle (Adapted from UNEP (2013), Shao et al. (2011))

Full size image

In general, early warning systems could help to contribute to three main elements necessary to prepare for a dust event: (1) communication of the actual event occurring, (2) its projected intensity and (3) its projected physical and geographic impact. With the first element, the general awareness of the event is heightened. The second and third elements of projected impacts can help people determine how to prepare for the dust event by taking measures such as evacuating, seeking cover, sealing doors and windows and securing outdoor assets such as vehicles, farming and manufacturing equipment. Additional preparation for farmers could include harvesting all or part of a crop early, if there is enough time before the dust storm hits (Stefanski and Sivakumar 2009).

8.2 Physical Characteristics of Dust Storms

8.2.1 Dust Storm Frequency and Origination

Dust storms can occur naturally or be induced by anthropogenic activity. A dust storm occurs naturally when high winds blow over soils that are vulnerable to surface disturbance (Wilcox 2012). Areas where soils have dried out due to prolonged drought or sudden dryness, or where there are a significant amount of dried out lakebed sediments, are also prone to dust storms. A dust storm that originated due to dried out sediments around Laguna Mar Chiquita in Argentina is pictured in Fig. 8.2. Anthropogenic activities that can cause or increase the chances or intensity of a dust storm include clearing of land for agricultural or infrastructure development, overgrazing and poor agricultural practice (Stefanski and Sivakumar 2009).

Fig. 8.2
figure 2

A dust storm occurred on 29 July 2012 near Laguna Mar Chiquita due to low lake levels and exposed, dried out sediment (NASA MODIS Aqua image, NASA 2012; visualisation by author)

Full size image

In dust storm-prone regions, storms can collectively occur on more than 300 days a year such as in Japan (MoE 2008) or fewer than 30 days a yearFootnote 1 such as in the United States (NCDC/NOAA 2013). Approximately 2,000 teragrams (Tg) of dust is emitted into the atmosphere per year with land surfaces receiving 75 % of the dust that is redeposited and 25 % deposited in the ocean (Shao et al. 2011). The Sahara Desert region alone emits 500–1,000 Tg/year, making it the most significant regional contributor (Goudie 2009). Other dust-producing regions include the Middle East, the Taklamakan Desert in northwest China, southwest Asia, central/western Australia, the Etosha and Mkgadikgadi basins of southern Africa, the Salar de Uyuni (Bolivia) and the southwestern United States (NRL 2009; Washington et al. 2003; Shao et al. 2011). These dust-producing regions have minimal moisture saturation, as indicated by a high erodible fraction valueFootnote 2 (Fig. 8.3).

Fig. 8.3
figure 3

Global dust-producing regions based on land-use erodibility fraction identified by the United States Geological Survey (NRL 2009)

Full size image

8.3 Geographic and Climatic Implications

8.3.1 Transboundary Dust Travel

Dust storms not only impact their origin area, but also can impact land, water and people a great distance away where dust, and the particles it carries, finally settles. For example, dust particles originating in inner and southern Mongolia and northern China contribute to dust events in Japan, the Democratic People’s Republic of Korea, the Republic of Korea and the Taiwan Province of China, causing seasonal ‘yellow sands’ and muddy rains (Lee and Liu 2004; Kimura 2012a, b). In the southern hemisphere, dust storms originating in eastern Australian can carry dust particles across the Tasman Sea to New Zealand, significantly contributing to soil development and geochemical cycles (Marx et al. 2009). As a result, creating policies pertaining to dust storm mitigation (i.e. vegetation restoration efforts) or developing early warning communication can be challenging. Further, since climatic changes will not affect all regions of the world in the same way or to the same degree, its influences on dust storms and their travel patterns is relatively unknown, thereby projecting uncertainty on populations that experience dust storms.

One of the most pronounced transboundary sources of dust is the Sahara Desert. Dust originating in the Sahara is frequently carried extended distances – east across the Nile Delta region and Mediterranean Sea, west across the Atlantic Ocean to the Caribbean, the USA, parts of Central and South America and north over to Europe. Dust originating in the Sahara has been reported as reaching as far east as Turkey in April 2012 (Mühr et al. 2013). This dust transport can negatively affect air quality, causing health and visibility problems (Prasad et al. 2010), but can also positively contribute to rainforests in Central and South America (Sivakumar 2005). Additionally, dust from the Sahara can also affect cloud development and precipitation patterns. There is some evidence that shows dust from this area can suppress tropical cyclone development, but more research is needed to confirm the claim (NASA 2013d). Figure 8.4 shows a dust storm that originated in the Sahara Desert blowing over Cape Verde and the Atlantic Ocean. The corresponding global aerosol index (AI)Footnote 3 image and the dust aerosol optical thickness image show how far offshore and over the Atlantic Ocean dust was actually being carried.

Fig. 8.4
figure 4

(a) NASA/GMAO GEOS-5 0 hour forecast image of dust aerosol optical thickness from 30 July 2013 (NASA 2013b); (b) Image from NASA’s OMI on the Aura satellite from 30 July 2013 (NASA 2013e); (c) NASA MODIS Terra satellite image from 30 July 2013 showing dust from the Sahara Desert blowing over the Atlantic Ocean and Cape Verde (NASA 2013c; visualisation by author)

Full size image

8.4 Climatic and Seasonal Variations with Transboundary Dust

Some regions are substantial dust producers year round, but seasonal changes can influence the level of dust intensity in other regions. For example, Asia experiences the most number of dust storms during its spring season, with the number peaking in April (Yang et al. 2007). Increased rainfall, snowmelt and onset of drought can influence the severity of dust events. The AI of an area is one method of identifying seasonal fluctuations in dust concentration. AI images show the presence of different aerosols, primarily dust.

Much like seasonal variations, climatic periods can also influence the frequency and intensity of dust storms. Through a comprehensive study of different regions in Northern China over a period of the past 1,000 years, Yang et al. (2007) found that the eastern arid regions experienced an increase in dust events during cold-dry periods compared to warm-wet periods as evidenced by increased amounts of dust found in ice cores. Vegetation cover was found to be relatively minimal during cold-dry periods, creating ideal conditions for dust mobilisation. Semiarid regions were also found to follow the same general correlation with dust events and cold-dry conditions, but correlated more strongly with precipitation patterns. From this observation, Yang et al. (2007) concluded that long-term precipitation changes more heavily influenced dust events than temperature changes did. Similarly, Marx et al. (2009) found that rates of Australian dust deposition varied with La Niña/El Niño patterns in which wet phases contributed to the addition of fine sediment to dust source areas, increasing the amount of sediment at the source as these new depositions dry out during dry phases and are available for transport. Climatic variations such as these need to be studied more in-depth at a regional level to understand the future implications of alterations to temperature and precipitation patterns that could be brought on by climate change.

8.5 Use of Modelling and Forecasting Networks for Early Warning Detection

The transboundary movements and issues previously discussed warrant the need for global dust storm forecast networks to help predict the onset, duration and path of a dust storm for the protection of both humans and the environment. While some countries have been monitoring dust events for 50 years or more (Wang et al. 2008; Shao et al. 2011), others have not. Models can help to close the data gap that exists in long-term meteorological data collections. Forecasts, derived from models, help decision-makers and scientists gather information about where and how fast a storm is moving. Information can be used by officials to deliver early warning information and by scientists to study patterns, as well as by medical officials and policymakers who can use forecast information to aid in determining the role of dust storms in origin and transport patterns of pathogens and disease (Stefanski and Sivakumar 2009).

Both regional and global models are essential for creating accurate forecasts due to variations in dust events across the globe. For example, March to May 2012 was an extremely active year for dust storms over the northern tropical Atlantic Ocean, the Arabian Peninsula, the northern Indian Ocean and the United States (Benedetti et al. 2013). However, during the same period, China experienced fewer dust storms than normal (10 instead of 17) and northern China had the least number of dust days since 1961 (Zhang et al. 2013). Furthermore, on the western edge of China, Mongolia experienced frequent dust storms, even a few particular days of extensive damage across the country in April.

An example of a regional model is the CUACE/Dust (Chinese Unified Atmospheric Chemistry Environment for Dust) model, which is used to produce forecasts for up to 3 days and is considered a quality model for dust events in East Asia (Wang et al. 2008). The model also has the ability to provide scientists and forecasters with information to make more educated impact projections because the model can provide insight about the distribution of desert and semi-desert land cover types, soil grain size, soil moisture content, snow cover and land use (Zhou et al. 2008). Alternatively, the Dust Regional Atmospheric Model (DREAM), utilised by the Barcelona Supercomputing Centre (BSC) and the United States National Weather Service (NWS), applies a global model to different regions and most extensively in the Mediterranean, North Africa and Middle East regions (NASA 2013a). With open online access, the BSC offers animated cycles of forecasts, using the 8b version of the DREAM model, in 6-h intervals from real time to 72 h in the future, for four substantial dust-producing regions in the world (BSC 2012). Forecasts using models such as the DREAM help to not only identify the immediate location of a dust particle, but also to determine point (or points of origin) and the anticipated path of the storm.

Other methods of creating forecasts and early warning information include using Light Detection and Ranging (LiDAR) data which allows for real-time monitoring. For example, the Ministry of Environment in Japan uses LiDAR to measure the presence of yellow sands and distinguish dust particles from other pollutants. The information is then incorporated into regional dust and sand storm monitoring networks to warn citizens of incoming dust storms as opposed to haze caused by pollutants. Ground-based global networks such as the Global Atmosphere Watch Aerosol LiDAR Observation Network (GALION) also use LiDAR data from several different regions and help to create a more complete picture of dust movement and the global dust cycle (Shao et al. 2011).

8.6 Early Warning Information Dissemination

An early warning message is only as effective as its method of delivery and interpretation. There are many ways to communicate messages with today’s technology including text message alerts, national and regional news broadcasts, website updates, email alerts and broadcasted warning signals through small communities. (Methods and challenges of early warning communication are further discussed in later chapters.) Most commonly, weather networks observe dust conditions (Shao et al. 2011) and disseminate the information (Davidson et al. 2003). The World Meteorological Organisation (WMO) has operated one of the most extensive forecasting networks since 2007. As described on the WMO website, the WMO Sand and Dust Storm Warning Network (SDS-WAS) is comprised of two nodes covering four regions: (1) Africa, the Middle East and Europe and (2) Asia (WMO undated).

Technological advances have encouraged information dissemination through websites and cell phone text messages. For example, the website for the Asia/Central Pacific Regional Centre with the SDS-WAS has a colour index with accompanying explanations to describe the severity of an impending dust event and China, which is included in this region, uses this information to display symbols to the public indicating time until a dust storm will occur (Zhang et al. 2007; WMO 2013). Text message, or SMS, warnings can also be used to instantly alert widespread populations of impending danger. Since 2012, the US NWS has supported a free Wireless Emergency Alert (WEA) service that delivers alerts to WEA-enabled mobile phones. The alerts warn of many types of hazards, including dust storms, and also offer suggestions of how to handle the situation, such as to avoid travel (NWS and NOAA 2013). However, with both websites and text messaging, if a vulnerable region does not have supporting infrastructure for Internet and cell phone alerts, then alternative means of communicating early warning information need to be developed.

8.7 Future Outlook

To increase the effectiveness of current forecasting and early warning initiatives, and to make future ones successful, some steps need to be taken. Primarily, several areas of research (Table 8.1) need to be addressed. The use of meteorological data and remote sensing information from satellite imagery could be used to complete much of this research (Stefanski and Sivakumar 2009; Urban et al. 2009; Kimura 2012b). In addition, increased awareness about forecast availability and education pertaining to how to interpret a forecast is needed to better prepare vulnerable populations and work towards mitigating the effects of dust events.

Table 8.1 Research aspects needed for the development of future forecasting and early warning
Full size table

Efficient and effective early warning systems, used not only to broadcast safety information for local populations, but also to predict drought conditions to prevent human-induced dust storms, can become more mainstream for current and future dust events. Global systems are also needed to address transboundary dust movements and effects. With impending climate change and a growing global population, consensus exists that additional research to fully understand the life cycle and prevalence of dust storms is needed.