December 9, 2016
Dr. James Aber
Emporia State University
Dusty Gutierrez, Liz Hagenmaier and Sophia McCall
Grapple
With disaster management, remote sensing systems have been and continue to be a highly valued tool for disasters such as flooding, cyclones/hurricanes, drought, earthquakes and tsunamis. Increased frequency of disasters has prompted global awareness for reducing and mitigating the impacts of disasters through disaster management. The cycle of disaster management include the following phases: disaster preparedness, disaster prevention, disaster relief, rehabilitation, and reconstruction. Remote sensing via satellite was largely adoptable throughout the disaster management cycle due to large area of coverage, short time orbiting, and its cost effectiveness. It does have its limitations due to data accessibility, differing quality between developing and developed countries, and technological limitations.
​
Studies in the 1990's were the first of many studies emphasizing the importance and applications of remote sensing in disaster management. It is seen as the fastest means of acquiring data and has been used in the pre- and post-disaster phase, especially in preparedness/warning and response/monitoring stages. In recent years, the addition of data resources and improved data resolutions have caused a larger collaborative and efficient framework for use of remote sensing systems as tools in disaster management. Additionally, more than 20 countries have remote sensing satellites that expand the available data resources. (Bello and Aina, 2014)
Science is an intellectual activity. Scientists gather and examine information and organize it into meaningful patterns in an effort to understand how the natural world works. Remote sensing is the science of collecting information without being in direct physical contact of the object of study. The term “remote sensing” was first introduced in 1960, before then aerial photography was the term generally used.
The roots of remote sensing can be traced back to the first photograph ever taken. The first known photograph was taken in 1827, by a man named Joseph Nicephoce Niepce. The image was produced using a pewter plate and light sensitive materials. Niepce attempts at photography were called heliography. In 1851 a man named Scott Archer developed the process of coating glass plates with sensitized silver compounds, these plates were referred to as “wet plates”. Shortly after the “wet plate” technique was developed the first aerial photograph was captured. The aerial photograph was attempted in 1858 by Gaspard Felix Tournachon using a balloon tethered over the Bievre Valley. In 1879 George Eastman developed a formula for a gelatin emulsion that covered the dry-plates and developed a machine to coat the plates. This would eventually lead to the invention of rolled paper film. Eastmans company, Kodak, is synonymous with photography today. As the process of photography further developed so did the uses for it. Aerial imagery was used by the military as a reconnaissance tool, taking photographs from planes.
Remote sensing today is a culmination of radar which can be used to create digital elevation models (DEMs), LIDAR (Light Detection and Ranging) used to measure chemicals in the atmosphere and detecting heights of objects on the ground, and aerial photos which can be used to create topographic maps.
Disaster Management: Flood
Flooding can occur in many types such as river floods, flash floods, dam-break floods, and coastal floods. Aerial extent, magnitude, frequency, duration, flow velocity, and time of occurrence are all variables for flooding events. In most phases of flood disaster management, remote sensing satellite data has been used successfully and strategically. Due to extensive cloud cover, optical sensors offer a limited use while systems such as Synthetic Aperture Radar (SAR) and RADARSAT offer a more useful imagery. Based on studies, data acquired from remote sensing technologies can be applied broadly in flood disaster management for preparing hazard assessment maps, creating hydrological models, quantitative assessment of the soil, generating flood risk maps, and early warning (Kordzakhia et al., 2011).
Introduction: A Brief history of Remote Sensing
Evolution of Remote Sensing in Disaster Management
​
Remote Sensing in Flood Modeling
Flood modeling has become increasingly important due to the increasing frequency of flooding, triggered by climate change impacts. There is now a growing demand for reliable data and improved modeling for flooding risk and impacts. Recently, remote sensing and flood modeling have been strongly integrated due to the advances in SAR remote sensing techniques and high-performance computing. Schumann, et. al., 2009, reviewed the recent efforts for integration of remote sensing with hydraulic modeling. There are four main study areas of this integration:
"...(1) the retrieval and and modeling of flood hydrology information from remote sensing observations...(2) the use of these data to calibrate and validate hydrodynamic models, (3) the potential of remote sensing to understand and improve model structures, and (4) the usefulness of remote sensing data assimilation with models."
-Schumann, et. al., 2009
Flood Modeling Integration
Disaster Management: Fire
Wildfires have the potential to be beneficial and harmful, presenting a challenge for ecosystem management. Wildfires are a natural part of several ecosystems, controlling age, structure and species composition of vegetation, regulating insect and disease populations, influencing nutrient cycles, and stabilizing and determining habitats for wildlife. Wildfires pose a continuous threat to human lives and property, ecosystems, and natural resources. They can be especially destructive in ecosystems where fires are uncommon and unnatural. Climate change is creating hotter and drier conditions effectively increasing the amount of wildfires and the length of the wildfire season.
The danger and cost of wildfires is increasing due to the significant increase in homes and businesses that are located in or near wildfire-prone areas. Several hundred million hectares of forest and other vegetation types are estimated to burn annually throughout the world, consuming billions of tons of dry matter and effectively releasing emission compounds that affect the atmosphere and human health (Leblon).
​
According to a study from the Union of Concerned Scientists (UNC) for every 1.8°F increase in temperature, a majority of the western United States will experience a significant increase in the area burned by wildfires. However, temperatures in the western United States are projected to far exceed that estimate, increasing 2.5° F to 6.5°F, by mid-century. A product of heat-trapping emissions from human activities.
​
Remote sensing data for fire management includes maps of historical patterns of fire ignitions, locations of buildings and other infrastructure, vegetation type and fuel loads, fuel condition (e.g. moisture content), topography, fire vulnerability and impacts of fire of vegetation regrowth, erosion, and other environmental characteristics (Chuvieco, E., and E.S. Kasischke 2007).
​
The ability of satellites to observe active fires is based on two physical principles; nocturnal satellite imagery is able to detect the light fire produces using the visible wavelengths of the electromagnetic spectrum and the high temperature of the fires increases the radiance emitted in the middle infrared bands, 3.7 µm being the most suitable for fire detection (Cahoon et al., 1992; Elvidge, 2001).
Remote Sensing in Fire Mapping Applications
Remote sensing data provides a faster and more cost effective way to obtain consistent observations and frequent updates over large swaths of area, which can often be remote and inaccessible. There are multiple remote sensing wildfire applications available that can help effectively manage wildfire-prone areas.
​
LANDFIRE (LF) is a Landscape Fire and Resource Management Planning Tool is a mapping program which provides products that are designed to support management of vegetation, fire, and fuels across multiple borders. The comprehensive geospatial data provides potential problematic areas of existing vegetation, surface and canopy fuel. Some specific data that is provided includes environmental site potential, existing vegetation (types, canopy, and height), and fire regime classes. The format is provided through an interactive map server, which is displayed in 30-meter grid spatial resolution raster data sets. It is also possible to download directly from the LF site (http://www.landfire.gov/). The LF management tool is shared between the U.S. department of Agriculture (USDA) Forest Service and U.S. Department of the Interior.
​
Map showing five fire regime groups based on frequency and severity; I. Frequent (0-35 years), low severity, II. Frequent (0-35 years), stand replacement severity, III. 35-100+ years, mixed severity, IV. 35-100+ years, stand replacement severity, V. 20+ years, stand replacement severity
Map credit: LANDFIRE (http://www.landfire.gov/lf_applications.php#maps)
Risk can be defined as the possible occurrence of an adverse event that has the capability of causing physical harm or monetary loss. The magnitude of the event is a measure of the gravity of that risk. Wildfire risks are increasing in part due to hotter, drier conditions caused by climate change. This risk is further increased due to local decision makers and property owners developing in wildfire prone areas.
​
A term that is widely used in reference to forest fires is “fire danger” which is a process that is used to evaluate and integrate factors that determine the ease of a fire starting and spreading, the impacts on the surrounding area, and the range of difficulty that is associated with controlling and extinguishing the fire. The role of remote sensing in evaluating fire danger or risk focuses on the ability to generate data regarding vegetation type and density, the moisture content, variations in climate and the physiology of the region. This is commonly referred to as fuel mapping. Fuel mapping is carried out using medium and high-resolution sensors, using Lansat-TM and MSS data. There have been several fuel mapping classification schemes proposed, currently the best-known system was developed by the U.S. Forest seriv’s Northern Forest Fire Laboratory (NFFL). NFFL based the scheme on the types of vegetation covering the forest; herbaceous, shrub, dead leaves, slash residues and basal accumulation materials. The surface vegetation may be partially obscured by canopy hindering the ability to estimate the fuel hight. Radar and Lidar sensors are a possible alternative due to their ability to penetrate canopy. The availability of satellite imagery and geographic information make it possible for fire scientists to effectively evaluate fire risk (Chuvieco and E.S. Kasischke, 2006).
Flood Hazard Assessment
With high-resolution flood modeling becoming a more feasible option worldwide, flood risk assessment maps can be developed and impacts of river floods can be reduced. Recent attempts in large-scale flood modeling has resulted in the Global Flood Awareness System (GloFAS), managed under the European Union. Remote sensing data is paired with streamflow data and other model inputs to provide a global-scale flood forecasting system (Dottori, et. al., 2016).
Grapple
Disaster Management: Drought
Drought poses significant food and water security issues which may lead to economic problems and financial concerns. This takes place on a global level. Due to regional variability in the global water cycle meteorological drought is a process closely related to climatic circulation patterns. However, for the scope of this project mainly the droughts within the United States will be analyzed. Drought is one of the most costly natural disasters in the United States (Gu et al., 2008). According to National Centers for Environmental Information (NCEI) droughts have cost $220.3 billons of dollars from 1980 to 2016 in the United States. That is over a span of thirty-six years with a recorded twenty-three events. That means 6.1 billion dollars on average is lost per year to drought (NOAA.gov, 2016).
Droughts are largely classified into four groups including, agricultural (soil moisture deficiency), hydrological (deficit in runoff, groundwater, or total water storage), meteorological (lack of precipitation) and socioeconomic (water demand, supply and social response) droughts. Every type of drought can be associated with a sustained precipitation deficit. Though, different elements of the hydrologic cycle respond to droughts in a different manner (AghaKouchak et al.,2015).
A broad-scale biogeographical shift in vegetation production is projected in response to the alteration of precipitation and temperature associated with global climate change. Vegetation alteration has profound ecological impacts and is an important climate-ecosystem feedback through their alteration of water, carbon, and energy exchanges of the landmasses (Adams et al., 2013). A particularly concerning topic is the elevation of temperature effecting the increased number and severity of droughts by generating widespread vegetation alterations via woody plant mortality (Adams et al., 2013).
Remote Sensing in Drought Mapping Applications
Several approaches can be utilized in remote sensing for drought and evapotranspiration. Thermal Remote sensing of drought and evapotranspiration is where water lost to the atmosphere via evapotranspiration serves to cool the Earth’s surface. Land surface temperature derived from remote sensing data in the thermal infrared band (8-14 um) is an incredibly valuable diagnostic source of biospheric stress resulting from soil moisture deficiencies (Anderson and Kustas, 2008). However, this is not the only way to utilize remote sensing techniques to capture evapotranspiration, mapping evapotranspiration at high resolution with internalized calibration (METRIC) is a satellite-based image-processing model for calculating evapotranspiration as a residual of the surface energy balance (Allen et al. 2007). This process can provide accurate and dependable information that is determined remotely, over large spatial scales; evapotranspiration can be aggregated over space and time.
Soil moisture is an important variable in land surface hydrology. The amount of moisture in the soil has a direct effect on the hydraulic conductivity. To be able to predict amounts of runoff, recharge rate, evaporation and other variables hydrologists need an accurate measurement of volumetric soil moisture (Cashion et al, 2005). Assessing these variables aid in the development of meteorological forecasting, impact assessments, wetland delineation, flood management and hydrological models. Satellite based passive microwave sensors show an effective way to observe soil moisture conditions over large areas. Soil moisture content has been successfully recovered over the last two decades with active and passive microwave techniques at various wavebands (Yilmaz et al., 2008). The aforementioned accurate monitoring and assessing near-real time vegetation drought conditions within the United States could provide accurate, widespread views of weather, timely information for effective drought planning and mitigation and likely would reduce economic losses (Gu et al., 2008).
NDVI Obtained from Wikicommons (Public Domain).
Remote Sensing and Drought Risk
Vegetation indicators like NDVI which is the normalized difference between the near infrared (NIR) and visible red reflectance (Tucker, 1979) which is obtained from optical satellite sensors have been largely utilized to evaluate the impacts of droughts on ecosystems. Optical-based vegetation indicators afford precious information on vegetation response to climate variability. NDVI is responsive to variations in both the intracellular spaces in spongy mesophyll of plant leaves and the chlorophyll content. Greater values of NDVI reflect higher vigor and photosynthetic capacity of vegetation canopy lower NDVI values are reflective of vegetation stress which results in chlorophyll reductions and changes in the internal structure of leaves caused by wilting—which is a common occurrence in the lack of adequate amounts of water for plant uptake. However, there are some downfalls to vegetation indicators like NDVI, they are sensitive to atmospheric effects, aerosols, cloud cover, water vapors, and land cover conditions. This makes for more of an information overview providing primarily information on conditions at the top of the canopy.
Microwave-based vegetation monitoring provides information on “real-time” aboveground canopy density and biomass. Microwave sensors can penetrate into dense canopy and are less affected by atmospheric conditions. This can be done either in day or nighttime conditions (Shi et al., 2008). Collectively, microwave sensors provide a relatively long-term record for investigating the impact and relative importance of droughts on not only local stateside but global biomass and vegetation change. This can greatly impact our understanding of ecosystem responses to drought—in different geographical regions, with hydrological variations and growth rates. The Global Ecosystem Dynamics Investigation LIDAR which is a laser based instrument designed to create 3D analysis of Earth’s forests will offer a unique route to monitor forest biomass and improve estimation of carbon fluxes—while being able to help aid in the identification of variation biomass due to lack of precipitation or increased evapotranspiration (Dubayah et al., 2014).
Summary
Severe climatic or geophysical events including landslides, volcanic eruptions, cyclones, floods, droughts, and fires that threaten property or people are considered natural hazards. Of these hazards floods, fires and droughts have been discussed in the prior paragraphs for using remote sensing techniques to analysis natural hazards. When the aforementioned hazards destroy peoples livelihoods, and lives they are categorized as natural disasters. In 2015 alone natural disasters had a devastating impact on a global scale with a reported number of 376 natural disasters, which caused the deaths of 22,765 people and displaced 110.3 million victims (EM-DAT, 2015). To put 110.3 million people being displaced in perspective that is roughly thirty four percent of the United States population. This number is astronomical. The previous sections discussed how to utilize remote sensing techniques in order to help elevate suffrage caused by natural disasters such as drought, flood and fire.
​
Several different techniques can be utilized via remote sensing such as NDVI for vegetation index, LANDFIRE or MODIS for fire management, along with SAR and RADARSAT that can aid in flood management. These are just a few of the remote sensing techniques that can be utilized to prevent the previously mentioned devastation and aid in elevating suffering. It is apparent that the use of technology such as remote sensing can be an invaluable tool in aiding with prevention and crucial time sensitive decision-making information that could save countless lives and property. Along with the aid in helping to secure a better defense for people living in high-risk areas prone to natural disasters there can also be research that can help to identify biodiversity or ecological sensitive areas.
References
Before and after flooding images in late July 2010, caused by heavy monsoon rains in several regions of Pakistan, including the Khyber Pakhtunkhwa, Sindh, Punjab and parts of Baluchistan. This image was acquired by NASA Terra spacecraft on August 11, 2010. https://images.nasa.gov/#/details-PIA13337.html
Before and after images following Hurricane Katrina on the Louisiana and Mississippi coasts. The images were acquired by NASA Terra spacecraft on August 14 and August 30, 2005. https://images.nasa.gov/#/details-PIA04385.html
Artist’s conception of the Landsat Data Continuity Mission (LDCM)
Remote Sensing of Flood Extent and Stage
Aerial photography is regarded as the most reliable source of remotely sensed flood area and extent but it can be cost prohibitive. The use of satellite data has been found to be a cost-effective alternative but it has its limitations and uncertainty. Some limitations include the ability or inability to acquire flood information routinely and penetration of clouds and open water bodies. Microwave (i.e. radar) remote sensing offers a solution to the sensor penetration limitation. Active microwave imagery from SAR is identified as the only reliable source of information for monitoring floods on rivers less than one kilometer in width. Use of the visible and thermal bands is seen as the most useful. SAR can be used for nearly real-time data for flood inundation modeling, especially during a current flooding event.
​
There is inherent uncertainty with remotely acquired data, both SAR and aerial photography. Image and sensor distortions complicate the assessment of the data for use. The magnitude of much of the uncertainty is a function of spatial resolution.
Hydraulic models requires a combination of complex factors and information. Remote sensing may provide the flood extent and stage that can be used to build hydraulic models. Flood extent data is most commonly used for model calibration and evaluation. Remote sensing data, such as the high-resolution lidar data, can be used to compare with model predictions. Integrating remote sensing data and hydraulic models is an established robust approach to flood forecasting.
This image from NASA Aqua spacecraft shows how surface emissivity -- how efficiently Earth surface radiates heat -- changed in several regions of Pakistan over a 32-day period between July 11 pre-flood and August 12 post-flood. https://images.nasa.gov/#/details-PIA13342.html
Remote Sensing and Fire Risk Assessment
Using Moderate Resolution Imaging Spectroradiometer (MODIS) the USDA Forest Service provides additional fire monitoring through an Active Fire Mapping Program available online. MODIS is an instrument that is currently aboard NASA’s Terra and Aqua satellites (https://modis.gsfc.nasa.gov/about/). The MODIS instrument is used to monitor the Earths land ocean and atmosphere. The MODIS Active Fire Mapping Program provides almost at real-time geospatial overview of the current wildland fire situations on both a regional and national scale. The instruments are viewing Earths entire surface in one to two days and acquiring data in 36 spectral bands (https://fsapps.nwcg.gov/afm/about.php).
Adams, H. D., M. Guardiola-Claramonte, G. A. Barron-Gafford, J. C. Villegas, D. D. Breshears, C. B. Zou, P. A. Troch, and T. E. Huxman (2009), Temperature sensitivity of drought-induced tree mortality portends increased regional die-off under global-change-type drought, Proc. Nat. Acad. Sci., 106(17), 7063–7066, doi:10.1073/pnas.0901438106.
​
AghaKouchak, A.; Farahmand, A.; Melton, F. S.; Teixeira, J.; Anderson, M. C.; Wardlow, B. D.; and Hain, C. R., "Remote sensing of drought: Progress, challenges and opportunities" (2015). NASA Publications. Paper 151. hp://digitalcommons.unl.edu/nasapub/151
Allen, R., Tasumi, M., and Trezza, R. (2007). "Satellite-Based Energy Balance for Mapping Evapotranspiration with Internalized Calibration (METRIC)—Model." J. Irrig. Drain Eng., 10.1061/(ASCE)0733-9437(2007)133:4(380), 380-394.
Anderson, M. C., and W. P. Kustas (2008), Thermal remote sensing of drought and evapotranspiration, Eos Trans. AGU, 89(26), 233–234, doi:10.1029/2008EO260001.
Bello, O. M. and Y.A. Aina (2013), Satellite remote sensing as a tool in disaster management and sustainable development: towards a synergistic approach, Procedia Social and Behavioral Sciences, 120 (2014), 365-373
​
Cahoon, D. R.Jr., B. J. Stocks, J. L. Levine, W. R. Cofer III, and K. P. O'Neill (1992), Seasonal distribution of African savanna fires, Nature, 359, 812–815.
Cashion, J., V. Lakshmi, D. Bosch, and T. J. Jackson (2005), Microwave remote sensing of soil moisture: Evaluation of the TRMM microwave imager (TMI) satellite for the Little River Watershed Tifton, Georgia, J. Hydrol., 307(1), 242–253, doi:10.1016/j.jhydrol.2004.10.019.
​
Chuvieco, E., and E. S. Kasischke (2007), Remote sensing information for fire management and fire effects assessment,J.
Geophys. Res.,112, G01S90, doi:10.1029/2006JG000230.
​
Dottori, F., P. Salamon, A. Bianchi, L. Alfieri, F.A. Hirpa, and L. Feyen (2016), Development and evaluation of a framework for global flood hazard mapping, Advances in Water Resources, 94, 87-102
​
Dubayah, R., et al. (2014), The Global Ecosystem Dynamics Investigation (GEDI) lidar, ForestSAT2014 Open Conference System, 4-7 November 2014, Riva del Garda, Italy
​
Elvidge, C. D. (2001), DMSP-OLS estimation of tropical forest area impacted by surface fires in Roraima, Brazil: 1995 versus 1998, Int. J. Remote Sens., 22(14), 2661–2673.
​
Gu, Y., E. Hunt, B. Wardlow, J. B. Basara, J. F. Brown, and J. P. Verdin (2008), Evaluation of MODIS NDVI and NDWI for vegetation drought monitoring using Oklahoma Mesonet soil moisture data, Geophys. Res. Lett., 35, L22401, doi:10.1029/2008GL035772.
​
Kordzakhia, G., L. Shengelia, G. Tvauri, M. Tatishvili, and I. Mkurnalidze (2011), Remote sensing for early warning of natural meteorological and hydrological disasters and provision of transportation safety over the Black Sea in Georgia, Procedia Social and Behavioral Sciences, 19, 532-536
2013, LANDFIRE.US_130FRG, Wildland Fire Science, Earth Resources Observation and Science Center, U.S. Geological Survey,
LANDFIRE 2010 Fire Regime Groups, http://www.landfire.gov/NationalProductDescriptions12.php
LANDFIRE Program Retrieved December 09, 2016, from http://www.landfire.gov/index.php
MODIS Moderate Resolution Imaging Spectroradiometer. NASA, Web. 09 Dec. 2016. <https://modis.gsfc.nasa.gov/about/>.
Schumann, G., P. D. Bates, M. S. Horritt, P. Matgen, and F. Pappenberger (2009), Progress in integration of remote sensing– derived flood extent and stage data and hydraulic models, Rev. Geophys., 47, RG4001, doi:10.1029/2008RG000274.
Shi, J., T. Jackson, J. Tao, J. Du, R. Bindlish, L. Lu, and K. Chen (2008), Microwave vegetation indices for short vegetation covers from satellite passive microwave sensor AMSR-E, Remote Sens. Environ., 112(12), 4285–4300
​
Tucker, C. J. (1979), Red and photographic infrared linear combinations for monitoring vegetation, Remote Sens. Environ., 8(2), 127–150.
​
Yirdaw, S. Z., K. R. Snelgrove, and C. O. Agboma (2008), GRACE satellite observations of terrestrial moisture changes for drought characterization in the Canadian prairie, J. Hydrol., 356(1), 84–92.
A wildfire near the California coast burned for almost 11 days in late July and early August. The natural-color image on the right was captured by the MODIS instrument on NASA’s Aqua satellite. The image on the left was acquired on the Landsat 8 satellite, which combines shortwave infrared, near-infrared, and green light and penetrates the smoke to provide a clear view of the burn scar. http://landsat.visibleearth.nasa.gov/view.php?id=88483