1

Tran A., Goutard F., Chamaillé

L., Baghdadi

N., and Lo Seen

D., 2010. Remote sensing and avian 1

influenza: a review of satellite image processing methods for extracting key parameters for avian 2 influenza H5N1 virus survival in water. International Journal of Applied Earth Observation and 3 Geoinformation, Vol. 12, Issue 1, pp. 1-8. 4 5

6

International Journal of Applied Earth Observation and Geoinformation - Review paper 7

8

Title 9

Remote sensing and avian influenza: a review of image processing methods for extracting key 10

variables to determine avian influenza virus survival in water from Earth Observation 11

satellites 12

13

Author names and affiliations 14

Annelise Trana,b

, Flavie Goutardb, Lise Chamaillé

a, Nicolas Baghdadi

a, Danny Lo Seen

a 15

a - Territories, Environment, Remote Sensing and Spatial Information Joint Research Unit 16

(UMR TETIS), Maison de la Télédétection, 500 rue J.-F. Breton, 34093 Montpellier Cedex 5, 17

France 18

b - French Agricultural Research Center for International Development (CIRAD), Animal and 19

Integrated Risk Management Research Unit, Baillarguet Campus, 34398 Montpellier Cedex 20

5, France 21

e-mail addresses: 22

annelise.tran@cirad.fr 23

flavie.goutard@cirad.fr 24

lise.c@wanadoo.fr 25

nicolas.baghdadi@teledetection.fr 26

danny.lo_seen@cirad.fr 27

Corresponding author 28

2

Annelise Tran 1

Telephone: +33 4 67 54 87 36 2

Fax: +33 4 67 54 87 00 3

4

3

Abstract 1

Recent studies have highlighted the potential role of water in the transmission of Avian 2

Influenza (AI) viruses and the existence of often interacting parameters that determine the 3

survival rate of these viruses in water; the two main variables are temperature and salinity. 4

Remote sensing has been used to map and monitor water bodies for several decades. In this 5

paper, we review satellite image analysis methods used for water detection and 6

characterization, focusing on the main variables that influence AI virus survival in water. 7

Optical and radar imagery are useful for detecting water bodies at different spatial and 8

temporal scales. Methods to monitor the temperature of large water surfaces are also 9

available. Current methods for estimating other relevant water variables such as salinity, pH, 10

turbidity and water depth are not presently considered to be effective. 11

12

Keywords 13

Remote sensing, water, avian influenza, H5N1 virus 14

15

16

4

Introduction 1

Avian influenza (AI) is a viral infection of birds that occurs worldwide. Depending on their 2

level of virulence to chickens, avian influenza viruses are divided into two groups, low-3

pathogenicity avian influenza (LPAI) viruses and high-pathogenicity avian influenza (HPAI) 4

viruses (Alexander, 2000). Among them, the Asian HPAI H5N1 virus is the cause of extreme 5

losses in the poultry sector, since 2003 an average of 200 millions of birds died of the disease 6

or have been culled to control its spread, global loss for control, vaccination and biosecurity 7

was estimated in billions of US Dollars (Otte et al., 2008). And at the same time this virus is 8

of greatest concern for human health with regard to its potential to become the next pandemic 9

strain (FAO, 2007; WHO, 2006). Identified for the first time in 1996 in China, the virus had 10

spread unrelentingly over South Asia since 2003, progressing westward from the end of 2005 11

into Eurasia and then over Europe, to finally reach Africa at the beginning of 2006 (FAO, 12

2007). Currently, the number of human cases around the world has reached 403, among 13

which there have been 254 deaths (as of 27 January 2009) (WHO, 2009). This crisis not only 14

represents an important threat for human health, but it also represents many socio-economic 15

problems involved in the agriculture and poultry industries. Despite enforced sanitary 16

measures, some countries (particularly in South East Asia) remain subject to periodic 17

recurrences of the virus and thus require new strategies to be implemented. In particular, to 18

improve disease surveillance, we need to have a better grasp on the ecology of the virus, 19

especially with respect to factors that favor the resurgence of new cases. 20

Several species of wild birds and poultry (notably ducks for the H5N1) have been identified 21

as reservoirs for avian influenza viruses (AIV) (FAO, 2007). The disease can be transmitted 22

either by direct contact with an infected host or indirectly through exposure to materials 23

contaminated with infected feces or respiratory secretions. Several experimental works have 24

been done to determine the environmental factors that influence the persistence of AIV in 25

5

feces. Beard et al. (1984) established that the virus could survive up to 35 days in wet feces 1

kept in closed vial at 4°C but this survival time will drop to 2 days if stored at 25°C. At 2

ambient temperature the survival of AIV virus in chicken manure was evaluated at 6 days 3

(Lu, 2003). The direct exposition of fecal material to UV light doesn’t have an impact on viral 4

infectivity (Chumpolbanchorn et al., 2006). 5

Transmission of AIV within wild bird communities via the ingestion of feces-contaminated 6

water is well documented (Brown et al., 2007). In late 70s , early 80s, wide varieties of AIV 7

were isolated from unconcentrated water from 6 lakes in Canada (Hinshaw et al., 1979), lakes 8

in United States (Halvorson et al., 1983) or from water ponds from duck farms in Hong Kong 9

(Markwell and Shortridge, 1982), showing that contaminated water supplies can constitute a 10

reservoir maintaining the virus in the environment and be a source for the contamination of 11

wild birds or other animal species. 12

This hypothesis is supported by several studies showing that the viruses are able to survive in 13

water for longer or shorter time periods depending on environmental characteristics (Brown et 14

al., 2008; Brown et al., 2007; Stallknecht et al., 1990; Weber and Stilianakis, 2008; WHO, 15

2007; Zarkov, 2006). Using this virus characteristic, researchers in Hong-Kong proposed 16

some surveillance strategy on live-birds market by sampling drinking water taken from 17

poultry cage, showing that the virus could survive between 8 to 24h in this environment 18

depending on level of chlorination and organic content of the water (Leung et al., 2007). 19

Recently a large diversity of AIV was detected in sediments of ponds in Alaska, hosting a 20

wide number of migratory waterfowl. Even if the study didn’t show the viability of the virus 21

in this condition they indicated that these sediments could play important role as year-to-year 22

reservoir (Lang et al., 2008). 23

Late work on mathematical model to simulate the transmission and the maintenance of AIV in 24

wild birds communities have shown that persistence of AIV in aquatic ecosystems is a key 25

6

parameters in the dynamics of infection in bird communities (Breban et al., 2009; Roche et 1

al., 2009). These results are underlying the importance to develop new methodologies to 2

identify potential risk areas of aquatic transmission. 3

Remote sensing technologies, which allow the mapping of environmental variables, have 4

been used in different epidemiologic studies (reviews by (Curran et al., 2000; Hay et al., 5

1997; Herbreteau et al., 2007; Kalluri et al., 2007)). 6

So far, remote sensing has been rarely used in the context of avian influenza. Few studies are 7

available that include the extraction of agricultural indicators (Gilbert et al., 2007; Gilbert et 8

al., 2008) or environmental indicators like surface temperature (Guo et al., 2005; Xiao et al., 9

2007). Gilbert et al. (2007) highlighted the strong association between intensive rice cropping 10

areas and the abundance of free-grazing ducks, using data from the Moderate Resolution 11

Imaging Spectroradiometer (MODIS). The latter appears to be a significant factor in the 12

transmission and persistence of H5N1 in Thailand (Gilbert et al., 2006). Land surface 13

temperatures derived from MODIS thermal imagery were used by Xiao et al. (2007) to 14

delineate the traveling wave of frost that triggers bird migration in the western Paleartic 15

region and thus impacts the H5N1 HPAI spread and seasonal dynamics. 16

All of these studies concern the detection of indicators for the presence of hosts (poultry or 17

wild birds) as risk factors for H5N1 HPAI transmission. At this point, no study exists to our 18

knowledge that deals directly with the detection of water as a potential survival environment 19

for the H5N1 virus and/or as place of risk for virus transmission. 20

The objective of this paper is to review the potential contributions of remotely-sensed data 21

and image analysis methods for identifying water bodies and inundated areas and their 22

classifications according to various key variables, to define AI-related health risk indicators. 23

After presenting the key factors known to influence the AI virus survival in water, a review is 24

given of what could legitimately be expected from various sensors (Synthetic Aperture Radar 25

7

(SAR), thermal and optical) onboard present and forthcoming Earth Observation satellites, in 1

terms of their mapping capabilities for water surfaces and related characteristics that influence 2

virus persistence. 3

1. Environmental factors influencing the survival of AI viruses in water 4

Fecal-oral transmission by contaminated water is a recognized mechanism of AI virus 5

transmission (Brown et al., 2007). This has spurred interest in the parameters affecting the 6

survival of AI viruses in water. Because little is known about the persistence of subtype HPAI 7

H5N1 virus in the environment, we review the environmental factors influencing the survival 8

of both lowly pathogenic as well as highly pathogenic avian influenza viruses, given that 9

under the same conditions HPAI H5N1 viruses seem to have reduced fitness in water 10

compared to LPAI, as suggested by recent works on AI virus inactivation in water (Brown et 11

al., 2008; Brown et al., 2007; Weber and Stilianakis, 2008). 12

Temperature. Experimental studies on LPAI (Brown et al., 2008; Brown et al., 2007; 13

Stallknecht et al., 1990; Webster et al., 1978) as well as HPAI (Brown et al., 2007; Phuong, 14

2005) viruses indicate that temperature is a key factor for virus survival; the persistence of the 15

virus in water is strongly reduced by high temperatures. For example, Brown et al. (2007) 16

estimated that the time required for an initial H5N1 viral concentration in distilled water to be 17

reduced by 90% was between 16 and 26 days at 17 °C and between 4 and 5 days at 28 °C. 18

Salinity. In the same study (Brown et al., 2007), three salinity levels were also used. It was 19

shown that higher water salinity could reduce virus survival, and that the effect of salinity was 20

more prominent at lower temperatures. These results are consistent with previous studies on 21

LPAI viruses (Stallknecht et al., 1990; Zarkov, 2006). In a recent study, Brown et al. (2008) 22

showed that LPAI isolates have important individual variations in response to increases of 23

salinity in water. From their results, they concluded that some LPAI viruses will reach their 24

peak of infectivity in brackish water with salinity between 10,000 and 20,000 ppm. 25

8

pH. LPAI viruses were found to be stable within a pH range of 7 to 8.5 (Brown et al., 2008; 1

Stallknecht et al., 1990). These results were confirmed by observations of the infectivity of 2

LPAI viruses from different natural water types (Zarkov, 2006). 3

Turbidity. Very little quantitative information is available on the impact of water turbidity on 4

the survival of AI viruses. According to Phuong et al. (2005), the H5N1 virus survives longer 5

in water with organic material, but the presence of living microorganisms reduces the survival 6

time of influenza viruses in water (Zarkov, 2006). 7

Overall, temperature and salinity seem to be the two key variables influencing HPAI virus 8

survival in water. For instance, H5N1 can be expected to persist for several weeks in water 9

with low salinity and low temperature (Weber and Stilianakis, 2008). In the next two sections, 10

we review the capacity of remote sensing to (i) delineate water bodies and map flooded areas 11

and (ii) extract those parameters influencing virus survival in water. 12

2. Mapping water bodies and flooded areas 13

Several methods are used in radar and optical remote sensing to delineate water bodies and 14

map flooded areas. This review covers sensors with medium to very high spatial resolutions, 15

which are compatible with the size of the water bodies of interest in the context of AI. Tables 16

1 and 2 summarize the characteristics of currently available optical and radar sensors to detect 17

water bodies. 18

Open water detection. Within the visible (VIS) and infrared (IR) spectra, water can be 19

characterized by a high reflectance in the blue wavelength, which rapidly diminishes in the 20

visible wavelengths to become very weak in the near infrared (NIR). Based on this spectral 21

signature, several indices were developed to delineate water bodies using ratios of spectral 22

bands; the most frequent wavelengths used were NIR, Green (G) and Middle Infrared (MIR). 23

These spectral indices aim to maximize the difference of pixel values between the object 24

being studied (i.e., water bodies) and other surfaces. Table 3 summarizes the indices 25

9

encountered in scientific literature and indicates how they are calculated and in which context 1

they were originally used. 2

The interaction between the radar wave and the surface of water is mainly affected by surface 3

roughness and depends on instrument parameters such as radar wavelength, incidence angle 4

and polarization. With SAR data in single polarization, it is possible to effectively detect open 5

water (Smith, 1997), when using the appropriate configuration. Baghdadi et al. (2007) 6

showed that the optimum configuration to discriminate water and land corresponds to SAR 7

data acquired at a high incidence angle (>30°), independently of polarization (HH, HV, VH, 8

VV) or to data acquired at a low incidence angle in cross-polarization (HV, VH), using 9

ASAR/ENVISAT data (C-band). 10

Detection of flooded areas. In addition to open water areas, the detection of flooded areas 11

with vegetation, like rice paddies, swamps or flooded forests, may be of interest in the context 12

of AI as those areas may be at risk for the transmission of the virus and act as natural 13

reservoirs for virus survival. In a recent study on environmental risk factors in Romania 14

(Ward et al., 2008), the presence of flooded shrub or herbaceous land cover was highly 15

associated with an increased risk of HPAI outbreaks in domestic poultry. 16

The use of optical remote sensing to map such flooded areas is limited when trees and floating 17

vegetation screen the water surface (Smith, 1997). Thus, some methods were developed using 18

optical sensors to monitor flooded areas, based mainly on time series analysis of multispectral 19

images and most often through the study of the temporal dynamics of vegetation indices 20

(Sakamoto et al., 2007; Xiao et al., 2005) or with land use change detection algorithms (Zhan 21

et al., 2002). 22

In radar remote sensing, the use of multiple frequencies is interesting for identifying several 23

types of inundated land (Smith, 1997). For example, Hess et al. (1995) showed that it was 24

possible to map inundated areas and flooded vegetation in the Amazon Basin using C- and L-25

10

bands SAR data. Baghdadi et al. (2001) used a decision tree based on the threshold of C-band 1

backscatter values in different polarizations to discriminate wetlands. Time series SAR 2

images were used in other studies to monitor flood dynamics (Martinez and Le Toan, 2007), 3

and to identify rice crops (Chakraborty and Panigrahy, 2000; LeToan et al., 1997; Panigrahy 4

et al., 1999; Shao et al., 2001) or marshes (Pope et al., 1997) through the analysis of changes 5

in backscatter values. 6

Finally, some authors demonstrated the ability of optical and radar remote sensing data fusion 7

to improve the classification of flooded land cover types (Castañeda and Ducrot, 2008; Oguro 8

et al., 2001; Seiler et al., 2008). 9

Due to the double-bounce scattering between the water surface and the vertical stems and 10

leaves of the vegetation, the radar returns from marshes or flooded vegetation are high and 11

easily separable from other surface types (Baghdadi et al., 2001). However, the detection of 12

flooded vegetation could be problematic if the vegetation cover is dense or if SAR data is 13

acquired using low radar wavelengths (X or C-bands). The use of high radar wavelengths (L-14

band) allows the radar waves to penetrate the vegetation cover and reach the underlying 15

water. 16

3. Water parameters assessment using remote sensing 17

The extraction of water parameters from satellite measurements has been the focus of 18

numerous studies. In this review, we focus on the key variables influencing the AI virus 19

survival in water as determined in section 2, i.e., temperature, salinity, pH and turbidity. We 20

also include water depth estimation using remote sensing, as this factor may influence 21

temperature. 22

Temperature. Water surface temperature can be estimated with remotely sensed thermal 23

infrared (TIR) images (Anding and Kauth, 1970; Handcock et al., 2006; Kay et al., 2005; Li 24

et al., 2001; Reinart and Reinhold, 2008; Schott et al., 2001; Wloczyk et al., 2006). Two 25

11

thermal bands are required to compensate atmospheric effects in an appropriate way (split-1

window technique)(McMillin and Crosby, 1984). Nevertheless, studies using one TIR band 2

from Landsat Enhanced Thematic Mapper (ETM+) imagery showed satisfactory results 3

(Schott et al., 2001; Wloczyk et al., 2006). Thus, depending on the spatial resolution of the 4

sensor (see Table 1), the temperature of rivers, lakes and seas may be assessed with good 5

agreement between remote sensing-estimated and ground-measured temperatures. Most of the 6

studies concern the estimation of sea-surface temperatures (Anding and Kauth, 1970; Franca 7

and Cracknell, 1994; Li et al., 2001); examples of applications to land-surface water bodies 8

(of interest in the AI context) are scarce, in part because of difficulty detecting streams and 9

small lakes using satellite remote sensing (Kay et al., 2005). For example, Reinard and 10

Reinhold (2008) used MODIS data to map temperatures in large lakes and Handcock et al. 11

(2006) estimated that reliable satellite TIR measurements are limited to large rivers (about 12

180 m across). For smaller rivers and streams, only airborne thermal remote sensing may be 13

used (Torgersen et al., 2001). 14

Salinity. The potential of passive microwave remote sensing (L-band, 1.4 GHz) has been 15

described in several studies because the natural emission from water is modulated by salinity 16

in the decametric wave range (Blume et al., 1978; Kachan and Pimenov, 1997; Le Vine et al., 17

1998; Singh et al., 1990; Wilson et al., 2001). Nevertheless, this is presently limited to 18

airborne remote sensing; the definition of a satellite sensor dedicated to the measure of sea 19

surface salinity is in process (Le Vine et al., 2004). 20

pH. The pH of water is one of the characteristics of water which may influence the survival of 21

avian influenza viruses in water (Brown et al., 2008; Stallknecht et al., 1990). Nevertheless, as 22

far as we know, there are no studies dealing with the estimation of water pH using remotely 23

sensed data, although Ferrari previously mentioned the influence of pH on the 24

characterization of turbidity using remote sensing (1991). 25

12

Turbidity. Water turbidity is generally characterized by two main variables, the Chlorophyll-1

a concentration (Chl-a) (a measure of phytoplankton abundance) and the Suspended Sediment 2

concentration (SS). These are key variables for assessing water quality. There is an 3

abundance of scientific literature available on the assessment of these variables using optical 4

remote sensing (Marcus and Fonstad, 2008). Indeed, turbidity is detected relative to water 5

color, which can be characterized by reflectance values in the VIS and NIR spectra. Basically, 6

Chl-a increases reflectance values in the Green and NIR wavelengths while SS increases 7

those in the Red wavelengths (Seyhan and Dekker, 1986). Thus, empirical or semi-empirical 8

models were developed to establish the relationships between in-situ turbidity measures and 9

reflectance values or indices (band combinations). More sophisticated algorithms (e.g., neural 10

networks, integrated approach using Geographic Information Systems) were also used to 11

improve the quantitative predictive values (Chica-Olmo et al., 2004; Pozdnyakov et al., 12

2005). For example, Landsat Thematic Mapper (TM) and ETM+ or MODIS Terra images 13

were used to assess turbidity in lakes (Giardino et al., 2001; Koponen et al., 2004; Wang et 14

al., 2006), marshes (Bustamante et al., 2008), large rivers (Aranuvachapun and Walling, 15

1988; Mertes et al., 1993; Stech et al., 2007), reservoirs (Verdin, 1985), estuaries and coastal 16

waters (Chen et al., 2007). The spatial resolution of the multispectral images setting limits the 17

size of the water bodies that can be monitored. 18

In one of the few examples of literature available on the use of SAR data to assess water 19

turbidity, Zhang et al. (2003) demonstrated that SAR was only marginally helpful for 20

improving the estimation of turbidity, as derived from Landsat TM data. 21

Water depth. Bathymetry can be assessed by optical remote sensing techniques based on the 22

relationship between water depth and water color, as light is attenuated through the water 23

column. Different methods were developed to map the depth of shallow water bodies, taking 24

into account that the signal measured by passive optical sensors is the result of interactions 25

13

between depth, turbidity and the bottom reflectance (Bierwirth et al., 1993; Kumar et al., 1

1997; Lyzenga, 1981). Nevertheless, air-borne remote sensing systems are required for 2

mapping depth in water bodies of small size (e.g.Leckie et al., 2005; Lejot et al., 2007; Lyon 3

et al., 1992). Some other techniques using airborne remote sensing are also operational: 4

stereophotogrammetry, ground penetrating radar and Lidar (Light detection and ranging) (see 5

(Feurer et al., 2008) for a review). On shallow seas, models assimilating radar images from 6

the European Remote Sensing Satellite (ERS) and a first guess depth map were developed by 7

Calkoen et al. (2001). 8

Discussion 9

Although scientific knowledge is still lacking on the global ecology of AI viruses in the 10

natural environment, different studies support the essential role of contaminated water bodies, 11

such as lakes and ponds, as environmental reservoirs of AI viruses. Moreover, flooded areas 12

may play a role in the amplification of virus transmission, as they are often zones of contact 13

between hosts (wild birds, domestic poultry). For these reasons, it seems important to be able 14

to map and characterize the water bodies to define risk indicators for the study, surveillance 15

and control of AI. The performances of Earth Observation Systems to achieve this goal were 16

reviewed in this paper, highlighting the potential and the limits of remote sensing for 17

identifying open water and flooded areas and the water parameters of importance for AI 18

viruses, i.e., temperature, salinity, pH, turbidity and water depth. 19

Both optical and radar imagery are effective for detecting water bodies. Despite the important 20

and frequent cloud cover in these regions, radar images seem particularly well-suited for 21

mapping flooded areas in tropical ecosystems, due to the ability of radar systems to acquire 22

information (Hess et al., 1995; Martinez and Le Toan, 2007). 23

Nevertheless, according to scientific literature, the characterization of the main water 24

variables seems to be a difficult task. Algorithms for estimating water-surface temperatures 25

14

from TIR sensors are effective only for large water surfaces (seas, lakes, large rivers). For 1

instance, SST and LST maps derived from MODIS data are produced and distributed 2

routinely (USGS, 2008). Water salinity could be measured using passive microwave remote 3

sensing, but these techniques are not yet operational. We did not find any studies concerning 4

the use of remote sensing to estimate the pH of water, whereas this variable may be of 5

importance for AI survival in water. There is abundant literature available on the assessment 6

of turbidity from space, based on the relationship between water color and Chl-a or SS 7

concentrations. Nevertheless, a calibration phase with in-situ measurements is required when 8

applying the methods to a new area. Finally, several attempts to map water depth have been 9

reported in the literature, but good agreement is obtained only in relative values, and it 10

remains difficult to obtain water depth accurately. 11

Most of the methods listed here are well-suited for large water surfaces, due to the spatial 12

resolution of the satellite imagery. Smaller water bodies like household ponds or small lakes 13

are also of great interest in the study of AI virus survival; in Cambodia, bathing in these types 14

of water bodies may be a risk factor for H5N1 HPAI transmission to humans (Vong et al., 15

2009). Instead, high or very high spatial resolution imagery should be used. However, the 16

adaptation of the methods to very high spatial resolution data may not always be possible, due 17

to the lack of appropriate wavebands and revisit characteristics. 18

In this review, we focused on the detection and characterization of water bodies, using Earth 19

Observation Systems, as there is some evidence for their role as reservoirs of AI virus when 20

contaminated (Brown et al., 2007; Weber and Stilianakis, 2008; WHO, 2007). In the future, 21

the capacity of remote sensing to map other AI relevant variables (soil moisture, texture, 22

vegetation cover, etc.) should be addressed, given that the knowledge of the ecology of AI 23

viruses in the natural environment indicates a link between AI risk and factors like the 24

characteristics of soils, vegetation, etc., which can be monitored using remotely sensed data. 25

15

Indeed, numerous others potential factors, like microflora, metal oxides, dissolved oxygen, 1

UV radiation, etc., affecting AI persistence in water are currently investigated (IRD, 2007; 2

European Commission, 2007). 3

In conclusion, remote sensing images can be used to map open water surfaces and flooded 4

areas and to derive a limited number of water parameters: temperature, turbidity and water 5

depth (the latter in relative values). In-situ measurements and/or additional modeling 6

approaches are required to assess others key parameters for virus survival such as pH or 7

salinity. Nevertheless, such preliminary approaches, as are currently available, can provide the 8

first classification of zones for potential AI virus survival in the water of a given area. To help 9

identify and map risk areas for AI, this information can be combined with other geographic 10

layers concerning the poultry trade, such as the density and structure of poultry farming, 11

migratory zones of wild birds, etc., using a Geographic Information System approach. 12

13

14

16

Acknowledgments 1

This research has been funded by the RIVERS project (Resistance of Influenza Viruses in 2

Environmental Reservoirs and Systems - http://www.manuguerra.eu/rivers/background.html) 3

of the 6th

framework program of the European Union, and the GRIPAVI project funded by the 4

French Ministry of Foreign affairs (http://avian-5

influenza.cirad.fr/fr/projets/recherche/gripavi). 6

The authors thank Jan Radomski and Franek Rakowski (Interdisciplinary Centre for 7

Mathematical and Computational Modelling, Poland), Sirenda Vong (Pasteur Institute, 8

Cambodia), Valéria Soti (CIRAD, nev-antropic, France), David Sheeren (Ecole Nationale 9

Supérieure Agronomique de Toulouse, France) for collaborations and discussions on this 10

topic, and Stéphane Dupuy (CIRAD) for his help on the review of sensors characteristics. 11

12

17

Tables 1

Table 1. Characteristics of current optical sensors 2

Table 2. Characteristics of current SAR sensors 3

Table 3. Spectral indices used for water detection 4

5

18

References 1

Alexander, D. J., 2000. A review of avian influenza in different bird species. Veterinary 2

Microbiology 74, 3-13. 3

Anding, D. and Kauth, R., 1970. Estimation of sea-surface temperature from space. Remote 4

Sensing of Environment 1, 217-220. 5

Aranuvachapun, S. and Walling, D. E., 1988. Landsat-MSS radiance as a measure of 6

suspended sediment in the lower Yellow river (Hwang Ho). Remote Sensing of Environment 7

25, 145-165. 8

Baghdadi, N., Bernier, M., Gauthier, R. and Neeson, I., 2001. Evaluation of C-band SAR data 9

for wetlands mapping. International Journal of Remote Sensing 22, 71-88. 10

Baghdadi, N., Pedreros, R., Lenotre, N., Dewez, T. and Paganini, M., 2007. Impact of 11

polarization and incidence of the ASAR sensor on coastline mapping: example of Gabon. 12

International Journal of Remote Sensing 28, 3841-3849. 13

Beard, C. W., Brugh, M. and Johnson, D. C., 1984. Laboratory studies with the Pennsylvania 14

avian influenza viruses (H5N2). Proceedings of the US Animal Health Association 88, 462-15

473. 16

Bierwirth, P. N., Lee, T. J. and Burne, R. V., 1993. Shallow Sea-Floor Reflectance and Water 17

Depth Derived by Unmixing Multispectral Imagery. Photogrammetric Engineering and 18

Remote Sensing 59, 331-338. 19

Blume, H. C., Ju, C., Kendall, B. M. and Fedors, J. C., 1978. Measurements of ocean 20

temperature and salinity via microwave radiometry. Boundary-Layer Meteorology 13, 295-21

308. 22

Breban, R., Drake, J. M., Stallknecht, D. E. and Rohani, P., 2009. The role of environmental 23

transmission in recurrent avian influenza epidemics. PLoS Comput Biol 5, e1000346. 24

19

Brown, J. D., Goekjian, G., Poulson, R., Valeika, S. and Stallknecht, D. E., 2008. Avian 1

influenza virus in water: Infectivity is dependent on pH, salinity and temperature. Veterinary 2

Microbiology, in press. 3

Brown, J. D., Swayne, D. E., Cooper, R. J., Burns, R. E. and Stallknecht, D. E., 2007. 4

Persistence of H5 and H7 avian influenza viruses in water. Avian Diseases 51, 285-289. 5

Bustamante, J., Pacios, F., Diaz-Delgado, R. and Aragones, D., 2008. Predictive models of 6

turbidity and water depth in the Doñana marshes using Landsat TM and ETM+ images. 7

Journal of Environmental Management, in press. 8

Calkoen, C. J., Hesselmans, G. H. F. M., Wensink, G. J. and Vogelzang, J., 2001. The 9

Bathymetry Assessment System: efficient depth mapping in shallow seas using radar images. 10

International Journal of Remote Sensing 22, 2973-2998. 11

Castañeda, C. and Ducrot, D., 2008. Land cover mapping of wetland areas in an agricultural 12

landscape using SAR and Landsat imagery. Journal of Environmental Management, in press. 13

Chakraborty, M. and Panigrahy, S., 2000. A processing and software system for rice crop 14

inventory using multi-date RADARSAT ScanSAR data. Isprs Journal of Photogrammetry and 15

Remote Sensing 55, 119-128. 16

Chen, Z. Q., Hu, C. M. and Muller-Karger, F., 2007. Monitoring turbidity in Tampa Bay 17

using MODIS/Aqua 250-m imagery. Remote Sensing of Environment 109, 207-220. 18

Chica-Olmo, M., Rodriguez, F., Abarca, F., Rigol-Sanchez, J. P., deMiguel, E., Gomez, J. A. 19

and Fernandez-Palacios, A., 2004. Integrated remote sensing and GIS techniques for 20

biogeochemical characterization of the Tinto-Odiel estuary system, SW Spain. Environmental 21

Geology 45, 834-842. 22

Chumpolbanchorn, K., Suemanotham, N., Siripara, N., Puyati, B. and Chaichoune, K., 2006. 23

The effect of temperature and UV light on infectivity of avian influenza virus (H5N1, Thai 24

20

field strain) in chicken fecal manure. Southeast Asian Journal of Tropical Medecine and 1

Public Health 37, 102-5. 2

Curran, P. J., Atkinson, P. M., Foody, G. M. and Milton, E. J., 2000. Linking remote sensing, 3

land cover and disease. Advances in parasitology 47, 37-80. 4

Franca, G. B. and Cracknell, A. P., 1994. Retrieval of Land and Sea-Surface Temperature 5

Using Noaa-11 Avhrr Data in North-Eastern Brazil. International Journal of Remote Sensing 6

15, 1695-1712. 7

Giardino, C., Pepe, M., Brivio, P. A., Ghezzi, P. and Zilioli, E., 2001. Detecting chlorophyll, 8

Secchi disk depth and surface temperature in a sub-alpine lake using Landsat imagery. 9

Science of the Total Environment 268, 19-29. 10

Gilbert, M., Chaitaweesub, P., Parakarnawongsa, T., Premashthira, S., Tiensin, T., 11

Kalpravidh, W., Wagner, H. and Slingenbergh, J., 2006. Free-grazing ducks and highly 12

pathogenic avian influenza, Thailand. Emerging Infectious Diseases 12, 227-234. 13

Gilbert, M., Xiao, X. M., Chaitaweesub, P., Kalpravidh, W., Premashthira, S., Boles, S. and 14

Slingenbergh, J., 2007. Avian influenza, domestic ducks and rice agriculture in Thailand. 15

Agriculture Ecosystems & Environment 119, 409-415. 16

Gilbert, M., Xiao, X. M., Pfeiffer, D. U., Epprecht, M., Boles, S., Czarnecki, C., 17

Chaitaweesub, P., Kalpravidh, W., Minh, P. Q., Otte, M. J., Martin, V. and Slingenbergh, J., 18

2008. Mapping H5N1 highly pathogenic avian influenza risk in Southeast Asia. Proceedings 19

of the National Academy of Sciences of the United States of America 105, 4769-4774. 20

Guo, J. P., Xue, Y., Cao, C. X., Cao, W. C., Zhong, S. B., Cai, G., Li, X. W. and Fang, L. Q., 21

2005. Study on the Highly Pathogenic Avian Influenza epidemic using Land Surface 22

Temperature from MODIS data. International Geoscience & Remote Sensing Symposium. 23

IEEE International Publication, 3599- 3602. 24

21

Halvorson, D., Karunakaran, D., Senne, D., Kelleher, C., Bailey, C., Abraham, A., Hinshaw, 1

V. and Newman, J., 1983. Epizootiology of avian influenza--simultaneous monitoring of 2

sentinel ducks and turkeys in Minnesota. Avian Dis 27, 77-85. 3

Handcock, R. N., Gillespie, A. R., Cherkauer, K. A., Kay, J. E., Burges, S. J. and Kampf, S. 4

K., 2006. Accuracy and uncertainty of thermal-infrared remote sensing of stream 5

temperatures at multiple spatial scales. Remote Sensing of Environment 100, 427-440. 6

Hay, S. I., Packer, M. J. and Rogers, D. J., 1997. The impact of remote sensing on the study 7

and control of invertebrate intermediate hosts and vectors for disease. International Journal of 8

Remote Sensing 18, 2899-2930. 9

Herbreteau, V., Salem, G., Souris, M., Hugot, J.-P. and Gonzalez, J.-P., 2007. Thirty years of 10

use and improvement of remote sensing, applied to epidemiology: From early promises to 11

lasting frustration. Health & Place 13, 400-403. 12

Hess, L. L., Melack, J. M., Filoso, S. and Wang, Y., 1995. Delineation of Inundated Area and 13

Vegetation Along the Amazon Floodplain with the Sir-C Synthetic-Aperture Radar. Ieee 14

Transactions on Geoscience and Remote Sensing 33, 896-904. 15

Hinshaw, V. S., Webster, R. G. and Turner, B., 1979. Water-bone transmission of influenza A 16

viruses? Intervirology 11, 66-8. 17

Kachan, M. V. and Pimenov, S. F., 1997. Remote sensing of water salinity at decameter 18

wavelengths. Ieee Transactions on Geoscience and Remote Sensing 35, 302-307. 19

Kalluri, S., Gilruth, P., Rogers, D. and Szczur, M., 2007. Surveillance of arthropod vector-20

borne infectious diseases using remote sensing techniques: A review. Plos Pathogens 3, 1361-21

1371. 22

Kay, J. E., Kampf, S. K., Handcock, R. N., Cherkauer, K. A., Gillespie, A. R. and Burges, S. 23

J., 2005. Accuracy of lake and stream temperatures estimated from thermal infrared images. 24

Journal of the American Water Resources Association 41, 1161-1175. 25

22

Koponen, S., Kallio, K., Pulliainen, J., Vepsalainen, J., Pyhalahti, T. and Hallikainen, M., 1

2004. Water quality classification of lakes using 250-m MODIS data. IEEE Transactions on 2

Geoscience and Remote Sensing Letters 1, 287-291. 3

Kumar, K. V., Palit, A. and Bhan, S. K., 1997. Bathymetric mapping in Rupnarayan-Hooghly 4

river confluence using Indian remote sensing satellite data. International Journal of Remote 5

Sensing 18, 2269-2270. 6

Lang, A. S., Kelly, A. and Runstadler, J. A., 2008. Prevalence and diversity of avian influenza 7

viruses in environmental reservoirs. J Gen Virol 89, 509-19. 8

Le Vine, D., Koblinsky, C., Pellerano, F., Lagerloef, G., Chao, Y., Yueh, S. and Wilson, W., 9

2004. A sensor to measure salinity in the open ocean from space. International Journal of 10

Remote Sensing 25, 1313-1318. 11

Le Vine, D. M., Kao, M., Garvine, R. and Sanders, T., 1998. Remote sensing of ocean 12

salinity: Results from the Delaware Coastal Current experiment. Journal of Atmospheric and 13

Oceanic Technology 15, 1478-1484. 14

Leckie, D. G., Cloney, E., Jay, C. and Paradine, D., 2005. Automated mapping of stream 15

features with high-resolution multispectral imagery: An example of the capabilities. 16

Photogrammetric Engineering and Remote Sensing 71, 145-155. 17

Lejot, J., Delacourt, C., Piegay, H., Fournier, T., Tremelo, M. L. and Allemand, P., 2007. 18

Very high spatial resolution imagery for channel bathymetry and topography from an 19

unmanned mapping controlled platform. Earth Surface Processes and Landforms 32, 1705-20

1725. 21

LeToan, T., Ribbes, F., Wang, L. F., Floury, N., Ding, K. H., Kong, J. A., Fujita, M. and 22

Kurosu, T., 1997. Rice crop mapping and monitoring using ERS-1 data based on experiment 23

and modeling results. Ieee Transactions on Geoscience and Remote Sensing 35, 41-56. 24

23

Leung, Y. H., Zhang, L. J., Chow, C. K., Tsang, C. L., Ng, C. F., Wong, C. K., Guan, Y. and 1

Peiris, J. S., 2007. Poultry drinking water used for avian influenza surveillance. Emerg Infect 2

Dis 13, 1380-2. 3

Li, X., Pichel, W., Clemente-Colon, P., Krasnopolsky, V. and Sapper, J., 2001. Validation of 4

coastal sea and lake surface temperature measurements derived from NOAA/AVHRR data. 5

International Journal of Remote Sensing 22, 1285-1303. 6

Lu, H., Castro, A. E., Pennick, K., Liu, J., Yang, Q., Dunn, P., Weinstock, D. and Henzler, D., 7

2003. Survival of avian influenza virus H7N2 in SPF chickens and their environments. Avian 8

Dis 47, 1015-21. 9

Lyon, J. G., Lunetta, R. S. and Williams, D. C., 1992. Airborne Multispectral Scanner Data 10

for Evaluating Bottom Sediment Types and Water Depths of the St Marys River, Michigan. 11

Photogrammetric Engineering and Remote Sensing 58, 951-956. 12

Lyzenga, D. R., 1981. Remote sensing of bottom reflectance and water attenuation parameters 13

in shallow water using aircraft and Landsat data. International Journal of Remote Sensing 10, 14

53-69. 15

Marcus, W. A. and Fonstad, M. A., 2008. Optical remote mapping of rivers at sub-meter 16

resolutions and watershed extents. Earth Surface Processes and Landforms 33, 4-24. 17

Markwell, D. D. and Shortridge, K. F., 1982. Possible waterborne transmission and 18

maintenance of influenza viruses in domestic ducks. Applied and Environmental 19

Microbiology 43, 110-5. 20

Martinez, J. M. and Le Toan, T., 2007. Mapping of flood dynamics and spatial distribution of 21

vegetation in the Amazon floodplain using multitemporal SAR data. Remote Sensing of 22

Environment 108, 209-223. 23

McMillin, L. M. and Crosby, D. S., 1984. Theory and validation of the multi window sea 24

surface temparature from space. Journal of Geophysical Research 89, 3655-3661. 25

24

Mertes, L. A. K., Smith, M. O. and Adams, J. B., 1993. Estimating Suspended Sediment 1

Concentrations in Surface Waters of the Amazon River Wetlands from Landsat Images. 2

Remote Sensing of Environment 43, 281-301. 3

Oguro, Y., Suga, Y., Takeuchi, S., Ogawa, M., Konishi, T. and Tsuchiya, K., 2001. 4

Comparison of SAR and optical sensor data for monitoring of rice plant around Hiroshima. 5

Calibration and Characterization of Satellite Sensors and Accuracy of Derived Physical 6

Parameters 28, 195-200. 7

Otte, J., Hinrichs, J., Rushton, J., Roland-Holst, D. and Zilberman, D., 2008. Impacts of 8

Avian Influenza virus on animal production in developing countries. CAB Reviews: 9

Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources 3, 1-18. 10

Panigrahy, S., Manjunath, K. R., Chakraborty, M., Kundu, N. and Parihar, J. S., 1999. 11

Evaluation of RADARSAT Standard Beam data for identification of potato and rice crops in 12

India. Isprs Journal of Photogrammetry and Remote Sensing 54, 254-262. 13

Phuong, D. Q. 2005. Seroprevalence study on avian influenza in rural poultry of Thai Binh 14

province and characterization of the environmental survival of the agents involved. MSc 15

Thesis, The Royal Veterinary and Agriculture University, Frederiksberg, Denmark. 16

Pope, K. O., Rejmankova, E., Paris, J. F. and Woodruff, R., 1997. Detecting seasonal flooding 17

cycles in marshes of the Yucatan Peninsula with SIR-C polarimetric radar imagery. Remote 18

Sensing of Environment 59, 157-166. 19

Pozdnyakov, D., Shuchman, R., Korosov, A. and Hatt, C., 2005. Operational algorithm for 20

the retrieval of water quality in the Great Lakes. Remote Sensing of Environment 97, 352-21

370. 22

Reinart, A. and Reinhold, M., 2008. Mapping surface temperature in large lakes with MODIS 23

data. Remote Sensing of Environment 112, 603-611. 24

25

Roche, B., Lebarbenchon, C., Gauthier-Clerc, M., Chang, C.-M., Thomas, F., Renaud, F., van 1

der Werf, S. and Guégan, J.-F., 2009. Water-borne transmission drives avian influenza 2

dynamics in wild birds: the case of the 2005- 2006 epidemics in the Camargue area. Infection, 3

Genetics and Evolution, in press. 4

Sakamoto, T., Van Nguyen, N., Kotera, A., Ohno, H., Ishitsuka, N. and Yokozawa, M., 2007. 5

Detecting temporal changes in the extent of annual flooding within the Cambodia and the 6

Vietnamese Mekong Delta from MODIS time-series imagery. Remote Sensing of 7

Environment 109, 295-313. 8

Schott, J. R., Barsi, J. A., Nordgren, B. L., Raqueno, N. G. and de Alwis, D., 2001. 9

Calibration of Landsat thermal data and application to water resource studies. Remote 10

Sensing of Environment 78, 108-117. 11

Seiler, R., Schmidt, J., Diallo, O. and Csaplovics, E., 2008. Flood monitoring in a semi-arid 12

environment using spatially high resolution radar and optical data. Journal of Environmental 13

Management, in press. 14

Seyhan, E. and Dekker, A., 1986. Application of remote sensing techniques for water quality 15

monitoring. Hydrological bulletin 20, 41-50. 16

Shao, Y., Fan, X. T., Liu, H., Xiao, J. H., Ross, S., Brisco, B., Brown, R. and Staples, G., 17

2001. Rice monitoring and production estimation using multitemporal RADARSAT. Remote 18

Sensing of Environment 76, 310-325. 19

Singh, R. P., Kumar, V. and Srivastav, S. K., 1990. Use of microwave remote sensing in 20

salinity estimation. International Journal of Remote Sensing 11, 321-330. 21

Smith, L. C., 1997. Satellite remote sensing of river inundation area, stage, and discharge: A 22

review. Hydrological Processes 11, 1427-1439. 23

26

Stallknecht, D. E., Kearney, M. T., Shane, S. M. and Zwank, P. J., 1990. Effects of pH, 1

temperature, and salinity on persistence of avian influenza viruses in water. Avian Dis 34, 2

412-8. 3

Stech, J., Alcantara, E., Novo, E., Shimabukuro, Y. and Barbosa, C., 2007. Turbidity in the 4

Amazon floodplain assessed through a spatial regression model applied to fraction images 5

derived from MODIS/Terra. Igarss: 2007 Ieee International Geoscience and Remote Sensing 6

Symposium, Vols 1-12 4550-4553 7

5334. 8

Torgersen, C. E., Faux, R. N., McIntosh, B. A., Poage, N. J. and Norton, D. J., 2001. Airborne 9

thermal remote sensing for water temperature assessment in rivers and streams. Remote 10

Sensing of Environment 76, 386-398. 11

US Geological Survey (USGS), Land Processes Distributed Active Archive Center (LP 12

DAAC), 2008. MODIS product table. Available at: 13

https://lpdaac.usgs.gov/lpdaac/products/modis_product_table (accessed on 23 December 14

2008). 15

Verdin, J. P., 1985. Monitoring water quality conditions in a large Western reservoir with 16

Landsat imagery. Photogrammetric Engineering and Remote Sensing 51, 343-353. 17

Vong, S., Ly, S., Van Kerkhove, M., Achenbach, J., Holl, D., Seng, H., Buchy, P., Uyeki, T., 18

Sok, T. and Katz, J., 2009. Risks of H5N1 infections from poultry to humans in Cambodia, 19

2006. Journal of Infectious Diseases, in press. 20

Wang, F., Han, L., Kung, H.-T. and Van Arsdale, R. B., 2006. Applications of Landsat-5 TM 21

imagery in assessing and mapping water quality in Reelfoot Lake, Tennessee. International 22

Journal of Remote Sensing 27, 5269-5283. 23

27

Ward, M. P., Maftei, D., Apostu, C. and Suru, A., 2008. Environmental and anthropogenic 1

risk factors for highly pathogenic avian influenza subtype H5N1 outbreaks in Romania, 2005-2

-2006. Veterinary Research Communications 32, 627-34. 3

Weber, T. P. and Stilianakis, N. I., 2008. Inactivation of influenza A viruses in the 4

environment and modes of transmission: A critical review. Journal of Infection 57, 361-373. 5

Webster, R. G., Yakhno, M., Hinshaw, V. S., Bean, W. J. and Murti, K. G., 1978. Intestinal 6

influenza: replication and characterization of influenza viruses in ducks. Virology 84, 268-78. 7

Wilson, W. J., Yueh, S. H., Li, F. K., Dinardo, S., Yi, C., Koblinsky, C., Lagerloef, G. and 8

Howden, S., 2001. Ocean surface salinity remote sensing with the JPL Passive/Active L-/S-9

band (PALS) microwave instrument. Igarss 2001: Scanning the Present and Resolving the 10

Future, Vols 1-7, Proceedings 937-939 11

3338. 12

Wloczyk, C., Richter, R., Borg, E. and Neubert, W., 2006. Sea and lake surface temperature 13

retrieval from Landsat thermal data in Northern Germany. International Journal of Remote 14

Sensing 27, 2489-2502. 15

Xiao, X. M., Boles, S., Liu, J. Y., Zhuang, D. F., Frolking, S., Li, C. S., Salas, W. and Moore, 16

B., 2005. Mapping paddy rice agriculture in southern China using multi-temporal MODIS 17

images. Remote Sensing of Environment 95, 480-492. 18

Xiao, X. M., Gilbert, M., Slingenbergh, J., Lei, F. and Boles, S., 2007. Remote sensing, 19

ecological variables, and wild bird migration related to outbreaks of highly pathogenic H5N1 20

avian influenza. Journal of Wildlife Diseases 43, S40-S46. 21

Zarkov, I. S., 2006. Survival of avian influenza viruses in filtered and natural surface waters 22

of different physical and chemical parameters. Revue Médecine Vétérinaire 157, 471-476. 23

28

Zhan, X., Sohlberg, R. A., Townshend, J. R. G., DiMiceli, C., Carroll, M. L., Eastman, J. C., 1

Hansen, M. C. and DeFries, R. S., 2002. Detection of land cover changes using MODIS 250 2

m data. Remote Sensing of Environment 83, 336-350. 3

Zhang, Y. Z., Pulliainen, J. T., Koponen, S. S. and Hallikainen, M. T., 2003. Water quality 4

retrievals from combined Landsat TM data and ERS-2 SAR data, in the Gulf of Finland. Ieee 5

Transactions on Geoscience and Remote Sensing 41, 622-629. 6

7

Web references 8

European Commision, 2007. RIVERS. Resistance of Influenza Viruses in Environmental 9

Reservoirs and Systems. Available at: http://ec.europa.eu/research/health/infectious-10

diseases/emerging-epidemics/projects/134_en.html (accessed on 19 May 2009). 11

Food and Agriculture Organization (FAO), 2007. Wild birds and avian influenza. Available 12

at: http://www.fao.org/docrep/010/a1521e/a1521e00.htm (accessed on 13 November 2008). 13

Institut de Recherche pour le Développement (IRD), 2007. Influenza aviaire et 14

environnement. Sciences au Sud, 42 :4. Available at : 15

http://www.ird.fr/fr/actualites/journal/42/sas_42.pdf (accessed on 19 May 2009) 16

US Geological Survey (USGS), Land Processes Distributed Active Archive Center (LP 17

DAAC), 2008. MODIS product table. Available at: 18

https://lpdaac.usgs.gov/lpdaac/products/modis_product_table (accessed on 23 December 19

2008). 20

World Health Organization (WHO), 2006. Avian influenza ("bird flu"). Available at: 21

http://www.who.int/mediacentre/factsheets/avian_influenza/en/ (accessed on 13 November 22

2008). 23

World Health Organization (WHO), 2007. Review of latest available evidence on potential 24

transmission of avian influenza (H5N1) through water and sewage and ways to reduce the 25

29

risks to human health. Available at: 1

http://www.who.int/water_sanitation_health/emerging/h5n1background.pdf (accessed on 14 2

November 2008). 3

World Health Organization (WHO), 2009. Cumulative Number of Confirmed Human Cases 4

of Avian Influenza A/(H5N1) Reported to WHO. Available at: 5

http://www.who.int/csr/disease/avian_influenza/country/cases_table_2008_09_10/en/index.ht6

ml (accessed on 27 January 2009). 7

8

30

Table 1. Characteristics of current optical sensors 1

Satellite Sensor Spectral resolution (µm)

[number and detail of bands*] Pixel size**

Revisit

period

Swath

width

Envisat MERIS 0.39 - 1.04 [15] Ocean: 1040 m x 1200 m

Land: 260 m x 300 m 1 – 3 days 1150 km

Terra MODIS 0.4 - 14.4 [VIS: 11; NIR: 6; SWIR: 3;

MWIR: 6; TIR: 10]

2 bands: 250 m (R, NIR)

5 bands: 500 m (VNIR)

29 bands: 1 km

1 – 2 days

2330

km

Terra ASTER 0.52 – 11.65 [VIS: 2; NIR :1; SWIR: 6;

TIR: 5]

15 m (VNIR)

30 m (SWIR)

90 m (TIR)

16 days 60 km

Landsat

5 / 7

TM /

ETM+ 0,45 – 12.5 [NVIS: 4; SWIR: 2; TIR: 1]

30 m (NVIS, SWIR)

60/120 m (TIR ETM+/TM) 16 days 180 km

Landsat7 ETM+ PAN: 0.52 - 0.90 15 m 16 days 180 km

EO-1 ALI

PAN: 0.48 - 0.69 10 m

16 days 37 km

0.13-2.35 [VIS: 4; NIR: 2; SWIR: 3] 30 m

EO-1 Hyperion 0.4 - 2.5 µm [220] 30 m 16 days 7.7 km

IRS-1C,

1D WIFS 0.62 - 0.86 [R, NIR] 188 m 5 days 806 km

IRS-1C,

1D LISS-3 0.52 – 0.86 [ VIS: 2; NIR : 1] 23 m 24-25 days 140 km

IRS-1C,

1D PAN PAN: 0.5 - 0.75 5.8 m 24-25 days 70 km

SPOT 4 HRVIR

0.5 – 1.75 [VIS:3; SWIR : 1] 20 m

26 days 60 km

PAN: 0,61 - 0,68 10 m

SPOT 5 HRG

0.5 µm – 1.75 [G, R, NIR, SWIR] 10 m (VIS)

20 m (SWIR) 26 days 60 km

PAN: 0,51 - 0,73 5 m

Ikonos

0.45 - 0.9 [VIS: 3; NIR : 1] 3.2 m

3 *** days 11 km

PAN: 0,49 - 0,90 0.82 m

Quickbird

0.45 - 0.90 [B, G, R, NIR] 2.44 m

3 *** days 16.5 km

PAN: 0,45 - 0,90 0.61 m

31

*VIS: Visible; NIR: Near infrared; VNIR: Visible and Near Infrared; SWIR: Short Wave Infrared; MWIR: Middle 1

Wave Infrared; TIR: Thermal Infrared; PAN: Panchromatic; B: Blue, G: Green, R: Red 2

** Medium, High and Very High spatial resolution data correspond to > 100 m, 10-100 m and < 10 m pixel size 3

respectively 4

*** off-nadir 5

6

32

Table 2. Characteristics of current SAR sensors 1

Satellite and sensor Band Incidence Polarization Spatial

resolution

Revisit

period

ERS-2 C 23° VV 12.5 m 35 days

RADARSAT-1 C 20°-50° HH 6.25 to 50 m 24 days

RADARSAT-2 C 20°-60° HH,HV,VH, VV 3 to 100 m < week

ENVISAT-ASAR C 15°-45° HH or VV or HH/HV or VH/VV or

HH/VV 12.5 to 75m < week

ALOS/PALSAR (3 modes :

fine, scansar, polarimetric) L 20°-60° HH or HH/HV or HH/HV/VH/VV 6.25 to 50 m -

Terra SAR-X X 20°-50° Single, dual, and full polarization 1 – 16 m < week

CosmoSky-Med X 20°-50° Single, dual, and full polarization 1 - 16 m < week

2

3

33

Table 3. Spectral indices used for water detection 1

Name Calculation Context Reference

NIR - Water detection (Work and Gilmer, 1976) (White, 1978) (Rundquist et al., 1987)

NDVI Normalized difference vegetation index

RNIR

RNIRNDVI

The index NDVI was originally used to

assess the biomass and vegetation primary production. Nevertheless, it can be of interest to detect water as NDVI shows positive values for vegetation, values close to zero for bare soil and negative values for water.

(Rouse et al., 1973) (Huete et al., 1997)

EVI Enhanced vegetation index 15.7.6

5.2

BRNIR

RNIREVI

EVI is derived from NDVI. EVI tends to limit

the aerosols effects and minimize soil effects.

(Huete et al., 2002)

NDWI Normalized difference water index NIRG

NIRGNDWI

NDWI is derived from NDVI, adapted to the

delimitation of water bodies with the use of reflectance in the green wavelength.

(McFeeters, 1996)

NDII Normalized Difference Infrared Index NDWI Normalized difference water index NDMI

Normalized difference moisture index LSWI Land surface Water Index

NDWINDII

NDMI

LSWI

MIRNIR

MIRNIR

Initially, Hardisky et al .(1983) showed a correlation between NDII and canopy water

content. The NDWI (Gao, 1996) and the NDMI

(Wilson et al. 2002) were used to detect leaves water content and to assess soil moisture. The LSWI (Xiao et al., 2005) is used to

detect soil moisture from MODIS data.

(Hardisky et al., 1983) (Gao, 1996) (Wilson and Sader, 2002) (Xiao et al., 2005)

MNDWI Modified normalised difference water index INH Normalized humidity index NDPI Normalized difference Pond index

INHMNDWI

NDPI

MIRG

MIRG

The MNDWI is derived from the NDWI

defined by Mcfeeters (1996) by the use of middle infrared instead of near-infrared (Xu, 2006). Water bodies are better delineated by a more efficient discrimination between open surface water and dry surfaces. The threshold of discrimination is located around 0. The INH was used by Clandillon et al.

(1995) in order to detect humidity in wetland environment. The NDPI was used for

detection of small ponds and streams semi-arid areas (Lacaux et al., 2007).

(Xu, 2006) (Clandillon et al., 1995) (Lacaux et al., 2007)

2

3

4

5

6