 Hello everyone, welcome to the next lecture in the topic passive microwave radiometry. So we are discussing the concepts about how a passive microwave radiometer will collect the signals from the ground, what those signal comprises of and for a pixel which what will be the total effect of the surface that will be reaching the atmosphere or the antenna like the surface temperature and the sky temperature combined together and what will be the brightness temperature of a mixed pixel which contains more than one feature. So, all these concepts we have been discussing. In today's lecture, we will get into the concept of signatures in passive microwave radiometry. So, signatures here I mean how different objects will behave in passive microwave radiometry and what essentially are we looking to do with this data ok. So, just take an analogy of passive microwave radiometry with thermal infrared remote sensing. In thermal infrared remote sensing our major aim will be to retrieve the surface temperature and use it for various applications. Our major aim will be that what is the temperature of the surface that is the true temperature what we call the radiometric temperature of the surface. We will try to calculate, but in microwave radiometry we will be measuring like a we will be measuring the brightness temperature of the surface that is a product of true temperature and its emissivity. So, let us take an antenna. So, the antenna will measure the combined effect of surface, atmosphere component that is reflected by the surface, direct emission by the atmosphere and some galactic effects say what is coming in from the cosmos. All these things will combine and produce one single temperature in the antenna. This in turn will also be affected by the antenna pattern. So, all these things we have seen. So, the final net temperature measured by the antenna or the antenna temperature comprises of the effects or includes the effect of surface, atmosphere, outer space and also the antenna pattern. What the satellite data providers will give us is they will take this antenna temperature, remove the effect of antenna pattern and atmosphere. They will like model the atmospheric effects try to remove the effect of atmosphere and they will give us the brightness temperature of the surface. So, normally if you download like a one product from any microwave radiometry data, we will have this brightness temperature of the surface. So, the brightness temperature is after removing the effect of say atmosphere antenna pattern. So, what we will have there effectively in the brightness temperature pixel, we will have the combined effect of emissivity and temperature of various features on the surface. We have also seen how they add up within a pixel if there are more than one feature, the net brightness temperature of that particular pixel is more or less equal to the area weighted average of brightness temperature of each and every feature. So, it is like a simple relationship which we can keep in mind and imagine it easily. So, from this brightness temperature, we will try to get meaningful data and use it for various applications. So, normally if you take example of usage of microwave passive microwave radiometry for soil moisture estimation and all. For some of the most of the applications using involving passive microwave radiometry, we will not be interested in calculating the temperature of the surface. For us to calculate the temperature of the surface, there is thermal infrared remote sensing. Here from the brightness temperature, we will try to get out or separate the effect of emissivity and temperature because the brightness temperature is a combination of emissivity and true temperature, right. What we will do is, we will try to remove this effect of temperature and see what is the net emissivity of that particular pixel. From that emissivity, we will try to infer some land surface information about land surface features. Say, what is the soil moisture there or what is the snow cover information there? For such land applications, normally we will be interested in calculating the emissivity of that particular pixel and using it for different things. So, here in passive microwave radiometry, for most of the applications, our aim will not be to calculate temperature. Our aim will be to separate the effect of temperature and emissivity from the brightness temperature and using that particular emissivity information how to estimate several land surface features. So, essentially we are interested here in seeing how emissivity varies due to land conditions of the land surface. So, this is like a major difference between what we do with the sensed temperature in tear remote sensing and passive microwave remote sensing. In tear remote sensing, our major aim is to calculate temperature for various applications. Sometimes we will also be interested in stopping with emissivity, but still for most of the applications, we will be interested to calculate the temperature of the surface. Here, we will try to, we will be more interested, I would not say completely, we will be more interested in knowing the emissivity of surface and by our knowledge of understanding how this emissivity varies with various land cover conditions, we will be able to estimate certain properties of land surface. So, this is like how we use the brightness temperature data in passive microwave radiometry. So, we will just take a look at variation of emissivity in tear domain. We have already seen it in the tear lectures, but just as a quick recap I would like to recall it. See, this is like a simple plot showing wavelength in x axis and emissivity of different features in the y axis. So, if you look at this 10 to 12 micrometer wavelength where we normally do our tear remote sensing, we can see the emissivity of various features on the earth surface varies with the limited amount. Say most of the emissivity are having a range of say 0.94 to 1 or we can even say 0.95 to 1. The variation is not much, just a 0.05 variation, not even like 1 it will be like 0.95 to 0.99 or something. No object is like a black body perfectly. So, the variation of emissivity is actually quite less in thermal infrared remote sensing between different features and the effect of temperature will be very high in the total radiance produced by the land surface. Say the just recall Planck's law. Planck's law tells us the total radiance emitted by an object is equal to product of emissivity into the Planck's equation times the temperature. That is the radiance L is given by emissivity times 2 hc square lambda power 5 exponential of hc by lambda kT minus 1. So, this is like the Planck's, this is the radiance that will be coming out from any particular object. Emissivity of different features does not vary much, temperature can vary. Say like if you take like within a pixel house, there may be house, there may be like a water body etcetera, water body may be at a different temperature, house may be at a different temperature and all these things will affect and produce like a combined radiance. So, the influence of temperature on the total radiance is pretty high in TIR remote sensing. So, it is always better to calculate temperature from this particular wavelength because whatever the radiation is produced, the effect of temperature is more in that in comparison with the effect of emissivity. Whereas, just let us look at in the passive microwave radiometry, the radiance produced is equal to a constant. So, here the constant is like 2 hc by lambda power 4 that is constant multiplied by emissivity into T. So, here the radiance produced is a direct product of emissivity and T. So, here the weightage that emissivity gets in controlling the radiance is pretty high, it almost has equal weightage with temperature and in addition to this, emissivity of features will vary a lot in passive microwave remote sensing. So, this suggests that if you want to get some properties of earth's surface, it is easy rather than as concentrating on temperature, it is easy for us to concentrate on this emissivity part and use it for understanding how emissivity varies with different land cover conditions. So, basically in microwave domain, the emissivity of a particular feature depends on the wavelength or frequency of our observation, the polarization in which we are observing, the look angle of the sensor 0 at nadir at some angle away from nadir etc. The T's 3 are basically the system properties or the sensor properties wavelength polarization and look angle all these are system properties. If you remove the system properties and if you look at the object properties, then the emissivity of the object will vary based on its dielectric constant. So, dielectric constant is it is like dealing with the electrical nature of an object. So, how good an object can conduct or not? So, based on the variation in dielectric constant, the emissivity of the objects will vary. If an object is a good conductor of electricity or if an object is a good conductor in general, its reflectance in microwave wavelengths will be very high and by virtue of this, the emissivity will go down. So, dielectric constant will influence emissivity to a large extent. Similarly, surface roughness whether a surface is smooth or rough that will influence the emissivity, the chemical composition of the surface and the temperature of the surface. So, the major two factors which play a role in controlling the emissivity of the surfaces is its dielectric constant and surface roughness. So, in most of the applications of passive microwave radiometry like applications here I mean application related to land, same estimation of soil moisture, ocean salinity and all these things. Scientists will be interested in finding out the relationship between this emissivity and dielectric constant and from the dielectric constant how to get the surface properties, whether it is soil moisture or ocean salinity or snow cover area whatever. So, just by looking at this from the brightness temperature by getting the emissivity information out we will be able to get a feel for or get information about the dielectric constant of the surface and by using this dielectric constant of the surface we will be able to retrieve several properties of the land or ocean surfaces. So, like I told like dielectric constant is like a major controlling factor. So, as the dielectric constant increases, the reflectance increases and emissivity decreases. Say for example, let us take dry soil and wet soil for dry soil the dielectric constant will be in the order of say 3 to 5 or 6 low dielectric constant it will have high emissivity in microwave wavelength. But if you add water to it and then the dielectric constant changes because water has a dielectric constant of about like 80. So, but soil has a dielectric constant of 3 to 5. So, that is like a huge difference. So, when water is added to the soil and when the soil gets wet the reflectance the dielectric constant changes completely the reflectance of microwave signals or the reflectance of soil in microwave wavelength increases and the emissivity decreases. So, this is like how emissivity changes with dielectric constant a very simple example. Just compare this with thermal infrared remote sensing. In thermal infrared remote sensing which is related to optical optical and thermal remote sensing goes together. Just recall how the reflectance of soil will change. We have studied in detail about how reflectance of vegetation, soil and everything will change. Just recall how reflectance of soil will change when water is added to the soil its reflectance will go down. When the reflectance goes down emissivity will increase emissivity is equal to 1 minus reflectance. So, wet soil will have high emissivity and low reflectance in optical and TAR wavelengths. In microwave wavelengths it is like exact opposite when water is added to soil due to change in dielectric constant the reflectance increases and the emissivity decreases. So, this is like a huge contrast between conventional optical and TAR remote sensing and microwave remote sensing. In microwave remote sensing the signals are mainly controlled by the electrical nature of the surface and also the physical nature of surface surface roughness the geometry and all those things will try to influence the signal that is emitted by the surface in passive microwave wavelengths. So, this is like though the concepts are similar we have to be careful in understanding how objects behave in different wavelength. Behavior of same objects say dry soil and wet soil are totally different when we observe in TAR wavelength or when we observe in microwave wavelengths. We have to be really cautious about it in which wavelength we are working on. Say this particular curve will tell us like how the emissivity of a surface changes. So, this is for sea water at 20 degrees Celsius at 35 GHz of observation. So, this is like angle to normal in degrees you can see that with change in angle with change in polarization this is V polarization that is H polarization the emissivity of the surface changes drastically. So, the emissivity varies very widely with respect to polarization and angle and similarly emissivity will also vary with different features and that is why because of this large variation in emissivity we will get a very large variation in brightness temperature in the microwave signals and using this information we will be able to get several other land surface properties. Say this table will give us say this is like microwave temperature of 3 different materials like theoretically calculated with the temperature the surface has a temperature of 300 Kelvin the atmosphere has a temperature of 40 Kelvin. So, assuming that atmospheric transmissivity is equal to 1 and everything like just neglecting the effects of atmosphere including only the surface effect atmospheric reflection on the surface and surface emission just including these 2 effects we are going to estimate how the brightness temperature changes. So, ground temperature is 300 Kelvin, atmospheric temperature is 40 Kelvin the sky temperature. So, this is like the dielectric constant of different features water solid rock and sand based on which there are equations to calculate reflectivity. So, that is from dielectric constant we can calculate reflectance roughly and then related with emissivity. So, here what we are doing is we are having information about reflectance for water the reflectance in microwave wavelength is roughly about 0.64. So, the emissivity is 0.36 here I rate emissivity for solid rock reflectance is 0.75 emissivity sorry 0.25 emissivity is 0.75 for sand reflectivity is 0.08 emissivity is 0.92. So, you can see the emissivity of the object changes drastically. In thermal infrared remote sensing in the 10 to 12 micrometre wavelength the emissivity of most of the features were in the range of 0.95 to 0.99 very less. In microwave just see these 3 the emissivity is like varying drastically 0.36 for water 0.75 for rock 0.92 for sand. So, these 3 features if they are present within 3 pixels let us say all these 3 features has same temperature 300 Kelvin. Then the net resultant microwave temperature of each pixel will be 134 Kelvin for water body, 235 Kelvin for solid rock, 280 Kelvin for sand. So, this tells us same brightness temperature, same atmospheric effect, but just by virtue of very large variation in emissivity the microwave brightness temperatures varies drastically. This suggests that the influence of emissivity on brightness temperature is very large. So, emissivity influences the brightness temperature of very large extent rather than the true temperature of object and hence in passive microwave radiometry our aim is to use this very large difference in emissivity between different features and from that emissivity difference calculate the dielectric constant of the objects. From the dielectric constant get it to different particular features say its emissivity or reflectance and use it for various applications. So, this is like a major difference in how we think and how we work in TIR remote sensing and passive microwave radiometry. The physical nature of the signal or the concept behind the signal emission is same, but the way in which we use the data is extremely different right. So, there is another major difference between TIR remote sensing and passive microwave remote sensing. In TIR remote sensing the temperature what we estimate we normally call it as surface radiometric temperature, but the other common name is skin temperature that is it indicates the temperature emitted by the surface is limited or whatever signals we get in TIR remote sensing is limited to very top millimeter or 2 of the surface. Whatever information we get it is just very thin layer of the top surface of whatever be there. So, if let us say it is like a bare soil field. So, whatever the signals we get in TIR remote sensing comes from maybe top 1 or 2 millimeters. Let us say there is like a very dense forest where we are seeing only the top of the canopy. You are not able to see the land surface itself. So, the temperature we are going to sense from the TIR radiometer is only the temperature of this dense canopy. So, the temperature we are sensing is essentially the skin temperature only the top 1 or 2 millimeters. In passive microwave radiometry the temperature signals or the radiance emitted by the surface will come from certain depth of the surface. Say in L band if you observe in L band most likely the signals coming out will contain information about first few centimeters of soil say 3 centimeters or 5 centimeters whereas in TIR remote sensing it is in order of millimeters 1 millimeter or 2 millimeters. So, there is like a huge difference in the depth from which the signals are going to come in passive microwave radiometry. So, in passive microwave radiometry we are going to get information about the surface. The emission that is being the emission is coming out from a significant depth of the surface in order of few centimeters. This depth will vary with respect to the wavelength in which we observe and also the moisture content of the surface. Increasing moisture content will decrease this penetration depth. Increasing or decreasing wavelength say L band to expand if you are moving in decreasing wavelength the penetration also will go down all these things are fine. But the whatever be the microwave wavelength the total penetration we have got will be almost higher than the penetration we will normally get in TIR remote sensing. So, the temperature what we collect in microwave signals they will not be known by the name of skin temperature we cannot call it because the signals essentially have originated from certain depth below the surface whatever be the feature but some amount of depth will influence the signal emitted and that will be reaching the sensor. So, this is again one of the major difference between what we sense in TIR remote sensing and passive microwave remote sensing. Now, we have discussed very broadly about what kind of signals will come to passive microwave radiometers what we will be interested on and so on. We will quickly see how the signals of different features will be in passive microwave wavelength. Say very similar to how we have discussed the spectral reflectance of vegetation, water, soil etc. Here also we will see the what factors will control the emission from vegetation, water and so on. First we will start with vegetation. So, the emission from vegetation in passive microwave radiometry is influenced by biomass, the geometric nature of the vegetation and also the vegetation water content within the canopy. Say normally like very dense vegetation if it is present it will have like a it will up the emissivity will be like healthy vegetation if it is there like full of water content emissivity will be lower it will appear like much cooler the brightness temperature will be pretty low. So, increasing biomass naturally indicates okay there is like a very large amount of vegetation there it may contain very good water content and hence will have like a very low temperature. If you compare this with like a dry bare surface it will dry bare surface will have high temperature or high brightness temperature I am talking about brightness temperature because of high emissivity okay. So, the emission of vegetation in microwave domain is influenced by the biomass, the geometric structure of vegetation canopy plus the vegetation water content. So, signals from soil and other components of vegetation will also penetrate the canopy and reach the sensor that is say if there is like a tree canopy here and if there is like a microwave radio I am going to locate that here not only the signal from this canopy will reach signals from this trunk signal from the soil underneath all will reach because of the penetration capacity of microwave signals the signal from the ground signal from the other non-canopy components also will reach it. So, this is like a major difference between passive microwave and thermal infrared remote sensing. The soil and other background effects reduces with actually it should be like decreasing wavelength. Sorry not increasing wavelength it should be with decreasing wavelength that is larger longer the wavelength say L band you are going to get like a information about objects present underneath that can be like soil you will you are going to get emissions from the soil also. But say you are looking at X band or something high frequency and low wavelength what you are going to get is only the information about the canopy and maybe like a small layer of the canopy. So, the soil and background effects will reduce when the frequency changes longer wavelengths will mean generally higher penetration. So, the emitted radiation from the from underneath the canopy gets attenuated strongly by the vegetation biomass and leaf water content. Say if the vegetation biomass is very large very high amount of vegetation and with very high water content then the signals from this particular soil or whatever that is underneath the canopy they will be attenuated by this attenuated means reduced significantly by this canopy and we will not be able to sense or get any information about the soil or whatever feature present underneath the canopy. So, this is how the signals in general will vary from vegetation. The next important feature which is which has very high amount of application of Pazimic radiometry soil. One of the major application for which Pazimic radiometry is used is estimating soil moisture. So, how wet or how dry a soil is? For bare soil soil moisture is a major influencing factor of emissivity because we have seen as an example that there is a huge difference in the emissivity of dry soil and wet soil like the dielectric content of dry soil is low, dielectric content of pure water is high when they mix together the reflectance of wet soil increases emissivity goes down. And this particular reason or this particular behavior is a major factor to use Pazimic radiometry for soil moisture estimation. In fact, the global soil moisture monitoring missions the SMAP and SMOS are effectively like dedicated missions for measuring soil moisture, ocean salinity and so on which are operating in L band. So, soil moisture is the major factor that affects the emission from bare soil. Increasing surface roughness will also increase the emissivity reducing the sensitivity of emissivity to soil moisture that is not only soil moisture but also surface roughness will also play a major role. Maybe we can just see with respect to example. Here we have two plots this first plot let us say this is plot A. So, here what we have seen is on the x axis we have the soil moisture in the first 2 centimeter of soil. On the y axis we have a normalized brightness temperature how the brightness temperature changes. All the brightness temperature are normalized between value of 0 to 1 just for explanation sake. We have 3 different field one is very rough means say the soil particles may be like very large or the soil may be drilled and like plowed and so on maybe like smooth surface is smooth surface of soil. We have 3 texture of soil plots each one is like a texture rough texture medium texture and smooth texture. Can see let us take smooth surface a smooth surface the slope of this line varies very steeply right the slope is very steep indicating that for a change for even like a small change in soil moisture there is like a significant change in the brightness temperature like as soil moisture increases the brightness temperature decreases that is the emissivity goes down basically right. So, here which suggests that the slope indicates basically the sensitivity larger the slope larger will be the change in soil moisture because of soil larger will be the change in brightness temperature because of the change in soil moisture right. So, for a smooth surface as soil moisture changes the brightness temperature changes to a significant extent. On the other hand for a rough surface the brightness temperature is always higher than the smooth surface. So, increasing roughness increases emissivity this is like one important point we have to remember increasing surface roughness increases the emissivity. So, this means rough surface whether it is maybe if it is like wet it will have high emissivity and high brightness temperature say let us say this particular point 40 percent bright 40 percent soil moisture. So, these two soils are at the same soil moisture content 40 percent but for a smooth soil the temperature the normalized brightness temperature is just above 0.5. For a rough surface the normalized brightness temperature is above 0.75. So, this suggests for a wet surface a change in surface roughness is going to increase the emissivity and increase the brightness temperature. So, this means if the surface is bare and smooth we will be able to get information about soil moisture in a better way. Soil moisture will influence the brightness temperature to a large extent. If the surface is rough the effect of surface roughness will add up which will try to increase the emissivity. We will get a mixed signal say let us say we do not know whether the soil is rough or not. Let us say we are we do not have any data we are going to get this particular point this particular brightness temperature. If we assume this if we do not have any other information we may assume the surface as smooth or rough whatever let us say by mistake we assume the surface as smooth like this may the signal is actually has come from a rough surface. But let us say we do not have that information by some thought we assume the surface as smooth. So, this will be the brightness temperature 0.75 and if we assume the surface as smooth then the 0.75 brightness temperature has occurred at around 20 percent soil moisture content. So, instead of calculating the moisture as 40 percent we will be calculating the moisture as 20 percent. So, this is like a very huge error 40 percent moisture 20 percent moisture we are making like 50 percent error in this 0.2 and 0.4 that is like a huge error right. So, when we try to estimate soil moisture from passive micro signals we should also know the surface roughness element how rough or how smooth the surface is that is a major factor. Because the surface roughness plays a major role for a dry soil if the soil is dry no issues rough surface and smooth surface will be at more or less close temperature you can see here at say around this 5 percent soil moisture content both the brightness temperature of all the 3 soils rough, medium and smooth or almost very similar. But as the moisture content increases the difference in brightness temperature becomes very huge and at very high moisture content definitely we should know whether the soil is smooth or rough ok. So, this is observed at L band 1.4 gigahertz. Similarly, let us have a look at the influence of frequency of microwave observation in estimating soil moisture. Same plot but here we have 3 lines where each line is at different different frequencies 10.7 gigahertz 5 gigahertz 1.4 gigahertz 1.4 gigahertz is L band 5 gigahertz is like C band and this can be I think around like X band ok. So, 3 different bands are there you can see the slope is the steepest for the L band. So, even like a small change in soil moisture has been seen or will produce a large change in the brightness temperature in L band when come when you compare this with 5 gigahertz or 10.7 gigahertz frequency. So, this suggests the sensitivity of brightness temperature to soil moisture varies with wavelength of observation and also the surface roughness. So, normally we normally we will get better information about soil moisture when we use higher wavelengths say L band or so. So, that is why the 2 missions map and smalls effectively they work in L band frequent they are L band radiometers like 1.4 gigahertz frequency. So, the signals coming out from soil is primarily influenced by soil moisture like especially bare soil. But the frequency in which we observe and also the surface roughness are also going to play a major role in influencing the signals that are influencing the emission from the soil surface. So, this is one of the major feature about passive microwave radiometry and this is this behavior of soil and influence of soil moisture on brightness temperature is used very widely for global soil moisture estimation. So, the next feature what we are going to see is like snow because snow also is one of like the major snow based cryospheric applications also like very widely done using passive microwave radiometry. So, this particular figures, so normally the microwave emission of snow depends on the snow wetness how wet or how dry snow is whether it is pure snow or mixed with water the snow particle radius and depth of snow. So, this particular figures give us how the emissivity of snow varies with x axis we have instant angle here in different plots we have different conditions of wetness average snow dry snow and wet snow say this dotted line is dry snow this broken dashed line is wet snow. So, for wet snow the emissivity is pretty high. For wet snow the emissivity is pretty high whereas for a dry snow the emissivity varies a lot the emissivity is pretty low and it varies a lot especially if you look at with respect to polarization say the one is vertical polarization one is horizontal polarization the emissivity is varying a lot whereas if the snow is wet the difference between vertical and horizontal polarization reduces drastically. So, as emissivity changes brightness temperature also will change. So, in general wet snow will have kind of like a very high emissivity and dry snow will have a very low emissivity relatively and also the difference between or difference of emissivity between V and H polarization will be largely seen for dry snow when we compare this with wet snow. Okay. So, this is like one of the important thing we should realize when we do snow remote something we are not going to see in detail each and every factor that controls emissivity of all the features but just to give a glimpse I am telling you all these things. So, the next feature is ocean how the ocean water emission will change one of the major applications of passive microwave radiometry in ocean remote sensing is to estimate ocean salinity like how salty sea water is salinity will influence the ocean circulation which has a global implications in global climate system. So, it is always but it is a need must to understand the ocean salinity and passive microwave radiometry produces or helps us to estimate this ocean salinity. So, the emission from the sea surface or sea water largely influenced by the temperature of the surface and also the salinity. So, in general increasing salinity decreases emissivity. So, that is what is given here and also on this particular plot you can see the sensitivity of brightness temperature with respect to salinity. Okay. So, here if you look at this particular line where I am putting this red dash you can see the y-axis has change in brightness temperature with respect to different properties of ocean. If you look only at salinity you can see at very low frequencies say in the order of less than 5 gigahertz there will be like a large change in brightness temperature for even like a small change in salinity because of this doh TB by doh salinity is very high. So, this suggests the change in salinity of ocean will produce a large change in brightness temperature that is the emissivity will basically decrease thereby reducing the brightness temperature. Okay. So, the sensitivity is pretty high and that is why low frequency say again the L band the 1.4 gigahertz band is widely used to estimate ocean salinity also. So, not only ocean salinity there are other features or other factors such as any presence of oil, sleek or wind everything will influence ocean emissivity. But as I said before our aim in this course is to not to explain all the variation all the features but just to give like a glimpse and also like one example how it can be used. So, just as like a very broad or very brief explanation we have seen how the emissivity of soil will change how the emissivity of vegetation will be influenced how the emissivity of snow will change etc. So, this is just to give you a glimpse of variation in brightness temperature or emissivity with different surface features. So, with this we conclude the topic of passive microwave radiometry. So, in passive microwave radiometry we have discussed in detail about or in brief about sorry in brief about the basic physical concepts behind passive microwave emission. What sort of what is like the detector we use like the antenna real aperture antenna synthetic aperture antenna the spatial resolution of passive microwave and sensors what kind of signals we get how those signals we use for our various applications and like a brief discussion about how the emissivity will vary for different features. With this we end this lecture and also this particular topic. Thank you very much.