 Welcome to the next lecture in the course remote sensing principles and applications. We are discussing the topic of thermal infrared remote sensing and in the last lecture we discussed in detail about the procedure to estimate or to retrieve land surface temperature from sensors with single thermal channel and our major concentration was retrieval of LST from Landsat series of satellites. So, just as a recap if we look at the way in which we derived LST is we first calculated radiance like dn to radiance conversion we have already seen how to do it in the previous lectures sometime before. From that radiance we have to do atmospheric correction like I introduced to you about like an online atmospheric correction tool which will help us to correct for the atmospheric effects. Apart from it we need to estimate surface emissivity in the spectral channel corresponding to Landsat sensors. Once we do this we substitute the terms in the general equation what is called the radiative transfer equation and then we can retrieve LST by inverting the plans function. As a brief summary we will just look at the slide. So, here in this particular slide this is the equation we used to for the retrieval of LST. So, the inverse plans function of the radiance received at the sensor after being corrected for atmospheric effect and surface emissivity effect. And atmospheric correction tool we have already seen and if we like substitute all the values like if we corrected satellite observe radiance and substitute in this sort of simplified plans function I call it as simplified plans function. This is specific for Landsat sensor like this constants k1 and k2 these will vary for sensor to sensor. So, this equation is specifically for like Landsat satellite with the corresponding k values for Landsat 7. Similarly, for other satellites such as Landsat 5 or Landsat 8 we have corresponding equations. So, once we have such the equation if we substitute the corrected radiance we will get this. So, this method what we have seen is known as retrieval of LST using radiative transfer equation like we took the basic equation like what all the components will go to the atmosphere we try to estimate them separately and we did it. Actually, given current circumstances current circumstances means even without access to any sophisticated models we will be able to retrieve Landsat phase temperature using this method because the atmospheric correction tool is available to us. Emissivity we can simply calculate using NDVI very simple method I have explained and substituting in this equation we can do this. But this is not the only way to retrieve Landsat phase temperature there are different ways to do it. One more popular way of retrieval of Landsat phase temperature from Landsat series of satellites is what is known as single channel method. That method we are not going to see in detail here, but the concept is rather than doing all these sort of corrections like atmosphere corrections separately for each and every component single channel method aims to simplify this like they will take only the primary components which will cause change in the sensor observed radiance most likely water vapor or CO2 concentration all these things and they will try to relate the radiance with respect to land surface temperature. If we know the water vapor content and other parameters that is developed by the single channel method we can just substitute them in those equations retrieve Landsat phase temperature. So, that method we are not going to see because for using that method we may have to access some atmospheric data at the time of satellite overpass or use some sort of atmospheric models retrieve data from such models and substitute but it is a one of the popular ways in which LST can be retrieved the single channel algorithm or single channel method. This is for sensors with one thermal channel that is why the name itself is either radiative transfer equation what we have seen or single channel method. For sensors with two or more than two thermal channels like even the reason Landsat 8 has two thermal channels within it that is band 10 and band 11 unfortunately we cannot use band 11 till now because of some artifacts but assume let us assume that we have two thermal sensors available to us in the same sensor say mode is has two thermal channels. In that case we can resort to a method known as the split window technique or the split window method. So, this in the split window method the general assumption is what you call the differential atmospheric absorption that is the two thermal channels they will normally be selected in a quanticus manner quanticus manner means say 10.4 to 11.5 11.5 to 12.5 though two thermal channels in the sensors most likely will be closely spaced with each other. So, assuming that atmospheric absorption will not vary much within the bands and emissivity will not vary within these bands and making similar sort of these assumptions people will develop equations relating the radiance with land surface temperature or the brightness temperature with land surface temperature whatever. So, the such equations once it is developed codes will be like people will use like high performance computers and atmospheric tools to refine this equations to a lot of extent. For example, one equation is given in this slide this equation is for a sensor called AVHRR which had like two bands if you see this equation you can this coefficients actually this TS is the land surface temperature A naught is a coefficient P and M are a coefficient. So, this is nothing but the coefficients are developed something like this these numbers epsilon of the emissivity delta epsilon is like the difference in emissivity and so on. And T4 T for AI are brightness temperature observed in the corresponding bands T4 T5 like band 4 and band 5 are thermal channels in AVHRR. So, this equation basically relates the brightness temperature observed from that particular sensor with the actual land surface temperature and brightness temperature what it told you it is just sensor absorbed radiance converted it into black body temperature. So, that will vary with respect to from band to band because of atmospheric artifacts and everything. So, but the final actual land surface temperature it should not vary in whatever band we look that is the physical concept. So, this brightness temperature is now related to land surface temperature using this coefficients. So, for with this particular equation as soon as we have surface emissivity we can just substitute them calculate this coefficients and get land surface temperature. So, this sort of equations relating radiance to LST or brightness temperature to LST it has been developed. And this is based on lot of assumptions behind it and lot of processing that has happened even before the sensor is launched. People will simulate the radiance using radiative transfer models people will simulate lot of atmospheric conditions they will try to see okay if this is the atmospheric condition if this is the surface condition what is the radiance there all possible combinations will be tried out and finally such equations will be derived. So, once such equations is derived as a end user we can use it simply and get the data or most likely when such equations are derived LST product itself will be generated by the satellite managers maybe like for modus NASA does it even for like India has this insert 3D sensor which is geostationary satellite the LST product is available from ISTO MOSDAC website we can download. So, such sort of products they will they it is easy to create such products when this split between two method is once the equations are established. But as end users or at the scope of this particular course we are not going to go deeper into how such equations are derived what are the physical concepts behind it all those things we are not going to see. But I just wanted to introduce you for sensors with more than one thermal channel such methods exist which enables the agency acquiring the data to process it into LST and give us give the users. So, now we can directly go and download LST data from modus we can directly go to MOSDAC website and download LST data from insert 3D satellite these things are possible we need not sit and retrieve LST because of these sort of methods. But for Landsat satellites as I told you before till now we have to retrieve LST and one simple method I have already explained to you which will be useful to you in several applications or in your other courses or other research activities. The next topic we are going to see is what is known as the thermal properties of terrain and how it will influence the observation of land surface temperature. Any material on earth surface if you take it will have certain characteristics with respect to heat energy or heat content. The first thing is heat capacity or thermal capacity. Let us say there is a material something is there. So, how much heat energy we should supply to that material in order to raise the temperature by 1 Kelvin or 1 degree Celsius. So, that amount of heat energy is the heat capacity of the material. If the material is able to withstand lot of heat energy its heat capacity will be pretty high. Some materials will quickly raise temperature even with a small amount of heating. Maybe like a very simple example is in our normal in our kitchens in our gas stove like in one burner heat water and heat milk and if we measure the rate of change of temperature between them those two liquids will have like will the temperature of both of the liquids will increase differently because of the difference in heat capacity of that particular liquids. Very similarly all the materials will have its own heat capacity and since materials can vary in mass in order to like kind of normalize it there is this specific heat capacity. So, specific heat capacity is the amount of heat required to raise the temperature of an object by 1 Kelvin like the or temperature of an object with 1 kg mass by 1 Kelvin. So, there we are talking about how much heat energy we should supply to raise the temperature of 1 kilogram mass of the material by 1 Kelvin. So, the material can be any weight 10 kilograms 20 kilograms whatever but how much heat we should provide for 1 kilogram that is specific heat capacity. Then the other important property is thermal conductivity. Thermal conductivity is simply put how fast heat will be conducted from one point to another point. Example let us take a one metallic rod if we heat one end of the metallic rod quickly the heat will be transferred to the other end and take like water put it in top of like a gas stove we supply heat to the at the bottom of the vessel it will be transferred to the water column like through convection. So, how much kind of how fast the heat moves from one point to another point thermal conductivity is but thermal conductivity is mostly dealing with solid but I gave water as an example. So, conduction process happens in solid basically and all these things combined together brings us to a more important property that we will that we will be interested in thermal remote sensing and that property is thermal inertia. So, what exactly inertia is we might have studied in our school physics that inertia is the inability of a body to change its state of motion or state of rest when it is when some force is exerted on it like when an object is kind of like moving it will resist coming to a sudden stop or if you are at rest it or if some material is at rest some force should be applied so that it starts moving. So, whatever be the object they will have some sort of inertia to resist this change of state similarly thermal inertia means the ability or the resistance offered by a body for change in temperature ok. So, like say you are trying to heat up some object continuously if the object has very high thermal inertia it will heat up very slowly the temperature will rise pretty slowly some objects will heat up very fast it has low thermal inertia. So, thermal inertia deals with the resistance of an object to change in temperature higher thermal inertia means lower the change in temperature lower thermal inertia means higher the change in temperature. So, this thermal inertia property depends on the specific heat capacity of an object the conductivity of an object and also the density of an object like mass by volume density. So, the thermal inertia depends on this thermal conductivity heat capacity and density of the material all these things will influence the inertia and higher the thermal inertia will lead to slower change in temperature of the object. But why do we need this property why we should worry about this thermal inertia anyway in remote sensing we are just sensing temperature. For some applications just sensing temperature at one time may be enough, but for some other applications like meteorological applications or whatever like sensing like land surface energy properties and all sometimes we may need to get the diurnal variation of land surface temperature. What exactly diurnal variation of land surface temperature is diurnal variation means variation of a certain property in a given day or with respect to time within the time span of one day that is 24 hours how the temperature changes. Actually this depends on surface property mainly like surface property and this kind of diurnal change will help us to infer some details about the earth surface maybe for energy balance studies maybe even like in olden days people were using this concept for geological mapping all those things. So, maybe we look at this thermal inertia property and this diurnal variation of temperature about few objects. So, this particular slide gives us the diurnal variation of surface temperature. So, this is from midnight of one day to midnight of next day. Let us take like one earths any part of the earth surface earths will be continuously emitting energy this is kind of like a non-stop thing as long as earth has some internal energy this radiation will be keep on going on or will be keep on going up towards the atmosphere. So, this is the long wave radiation what we are currently using for our thermal infrared studies in the right now we are talking about 8 to 14 micrometers only. But this radiation will be spanning across like from 3 to 5 micrometers to even like in microwave range. So, this radiation will be taking the energy away from the earth surface. So, when energy is being lost what will happen the temperature of the object will go down if no other source is provided. So, during night time when the sun's energy is absent earth will be keep on radiating and it is due to this continuous radiation the temperature will be slightly decreasing. So, it will be kind of here this is a surface temperature curve this will be decreasing because of continuous loss of energy. So, earth will reach some sort of minimum temperature just at like dawn when the sun is about to rise ok. So, when sun rises what will happen sun will provide the energy source. So, this is what to say this is earth's energy curve. So, it is continuously losing energy. So, this is by sun. So, till dawn no energy is there from sun after that sun's energy begins to rise solar energy will begin to come when solar energy starts to come and fall on the earth's surface what will happen this energy will add up to earth's surface and this will heat up the earth's surface thereby rising its temperature. So, after sunrise the temperature will begin to increase because there is energy surplus whatever the earth is radiating that is fine, but the incoming energy is more than that the earth's surface is being heated up by solar radiation. So, as solar radiation heats up the earth's surface the temperature will rise and this radiation will again increase because as temperature rises more radiation will go that is we have seen in Planck's law as increased temperature leads to increased radiation. But still during noon time solar radiation will peak energy will be like at surplus because of this energy surplus surface temperature will be continuously rising and at certain point in daytime it will reach a maximum state because after noon time solar radiation will go down right same thing. So, then but there is you can observe there is kind of like a time lag between where the energies is peaking and the temperature maximum temperature is attained because it needs some time for the earth to absorb this energy and increase its temperature. There is always a lag between peak energy supply of peak energy and peak temperature. So, what essentially happens is during daytime this external heat energy from sun increases the temperature of earth's surface features. Then afternoon again the solar radiation will begin to go down slowly it will decrease and after sunset it will become 0. So, similarly temperature also will begin to decrease because earth is continuously radiating energy. So, the temperature will begin to go down. So, this cycle increase in energy like continuous radiation decreasing temperature then increasing temperature then decrease. This cycle will happen continuously for all the objects this is the basic concept of diurnal variation of surface temperature or everything happens because of incoming solar radiation during daytime that is like the driving force. This incoming energy and or this incoming energy will change the temperature of all features on the earth's surface and different objects on the earth's surface will behave differently to this additional energy. So, that depends on thermal inertia of the objects. Now we will quickly see some diurnal temperature pattern a typical patterns for few earth's surface materials. See this is for bare soil or rock or concrete this dark black line this dotted line is for vegetation and this is for water. So, this kind of like uniform dotted line is for water then it is for moist bare soil all these things. So, what essentially it means different features behave differently to the solar radiation or change in temperature. Water has very high thermal inertia it actually resist tendency to change its temperature. So, the temperature variation if you take the difference between the maximum temperature in a day and minimum temperature in a day the difference will be quite less. On the other hand for bare soil or dry soil this is bare dry soil or concrete kind of materials they do not they will not have very high thermal inertia and they will change temperature pretty quickly. They will heat up fast the temperature will rise up very fast similarly during night time they will cool down really fast. So, during night if you observe here water has higher temperature. So, water will appear warmer than this vegetation or bare soil and all. But on during daytime bare soil will have higher temperature water will have lower temperature. So, the temperature will vary because during daytime bare soil and rocks will heat up extremely fast they will rise to very high temperature. Water will resist temperature change it will rise very slow. During night time again water will resist change in temperature it will try to remain at the same temperature. On the other hand dry soil and bare rocks will cool down very fast they will lose temperature lose energy. So, these things play a major role that when we do remote sensing especially thermal remote sensing and what do we observe. So, during night time like midnight and pre-dawn time if we take thermal infrared measurements water bodies may appear warmer than typical land surface. On the other hand during daytime water bodies will appear cooler than buildings or back surface this is due to the variation in thermal properties. So, higher thermal inertia lower will be the change in temperature. Here just look at two important points one point here and one point here just before or after sunrise like there will be and similarly just before kind of sunset. There will be like two points where temperature of all the materials will coincide or they will intersect which means at certain time point most likely very close to early morning sunrise time and similarly close to sunset time the temperature of various features will will be almost the same that is because at that point we have a thermal crossover. Thermal crossover means sun has already come it has just started to shine. So, rocks and other materials are going to or started beginning to rise in temperature water is still resisting. So, that certain time point all temperatures of like temperature of various features will be more or less the same. Same thing will happen in the evening after like around like sunset bare soil and all will lose temperature fast water will be still maintaining the temperature. So, what so bare soils temperature will come down water's temperature will come down slightly at some point they will meet. This kind of similar crossovers will happen and we are not supposed to do thermal infrared remote sensing at that particular time interval. Because our interest is to map the temperature of various features and these things will influence how we use it for various applications and observing or making thermal infrared observations at this particular point of time will not help us to get information about the temperature of the land surface we will get uniform temperature. So, normally at this thermal crossover point it is not suggested or not recommended to make thermal infrared measurements if our aim is to measure the temperature of various objects of surface. Same maybe in the while we discussed briefly about applications I will tell you about surface energy balance equation that is how we map evapotranspiration water lost to the atmosphere that basically depends on this temperature differences in the surface. One of the principles we use how much temperature is there. So, if we take what to say if we take thermal infrared observations during like this thermal crossover time we would not get much information about it. Similarly, for urban heat island studies it may not be of much use. So, if we are interested in obtaining the temperature of various objects it is not recommended to make tier observations at this thermal crossover time. So, before we conclude this particular topic I like to briefly introduce you about the directional anisotropy effects that exist in thermal infrared measurements. We have seen directional anisotropy effects in optical remote sensing like visible and air bands. Directional anisotropy means basically objects will look different when we look at from different different points in space that is the bi-directional reflectance distribution function BRDF all these concepts we have seen before. Even in thermal remote sensing even when we observe temperature this kind of directional anisotropy will come in because the first major reason is emissivity is directional. Most of the earth surface features will emit differently in different directions. This is a primary reason and also the application of heat to surface features may be different from different directions. It may be one small example I will tell you in this particular slide. So, here sun is here. So, this is like kind of like a row crop let us say some sort of crop like grapes or something is standing. So, these are like planted in rows. So, this will be crop one line of crop this will be soil another line of crop another line of soil like this it is there. So, sun is here. Let us say one sensor is looking from this direction one sensor is looking from this direction due to this solar radiation. So, it is coming in from this side. So, what will happen there will be like a thermal shadow here thermal shadow means this particular portion on the soil surface will not receive direct solar radiation. All the solar radiation will be blocked by this canopy here this part will not receive much radiation will remain cool itself like temperature will not raise a lot. So, same thing will happen to all the soils in between. Say some sensor is observing from this side the same side as where the sun is. So, this sensor will most likely see the sunlight portion of the canopy which is already at like experiencing direct sunlight. This will make this portion appear warmer the canopy. Similarly, let us say some other sensor is looking from here this particular direction. If it is seeing like the soil part it may observe like a mix of cool soil and shaded portion of this canopy the other portion of canopy sunlit this portion is shaded because of solar radiation. So, these sort of change in solar heating will influence which side of vegetation canopy is heated if there is a tall tree standing if sunrise if solar radiation is from this side one side of the tree will be heated up faster than the other side. Similarly, the other side the soil will be in shadow the temperature will be lower. So, if a sensor observes some nadir it will observe a mix of this warm side of tree the shaded side of tree and this shaded soil. If a sensor observes from the same side as sun it will observe the sunlit portion and warmer part of canopy. If the sensor observes from another side it will observe the shadowed portion of canopy. All these things will change the temperature that we are sensing. This complicates matters if you want to measure temperature for certain applications and if we measure from different different directions we may end up in measuring different temperatures. So, the directional anisotropy is not only a problem in optical remote sensing but it is also an issue in thermal infrared remote sensing especially when we need very high accurate temperature measurements the direction in which we look will influence the temperature readings. Example is given in this particular slide. So, this will will tell us clearly like the solar radiation is now the sun is shining from this particular point. So, where the sensor is seeing when now in this first particular case we are sensing a sparse canopy. Sparse canopy means there is vegetation is present there is like a large soil gap then another vegetation stand is present. So, this sensor most likely will see a mix of vegetation and soil giving rise to a complex signal. If the canopy is dense in case of like a dense forest all the signals may be coming in only from the canopy itself no signals from the ground may come in. So, the temperature what we measure may be different. Similarly, let us say the sun is here and the sensor is also looking in the same direction. So, the sensor is going to observe only the sunlit portion of this canopy and soil which is going to give a increased temperature. So, all these things whether the canopy is dense whether the canopy is not dense which side solar heating is happening all these things will influence the temperature that we are measuring. So, these things we have to keep in mind when we need accurate temperature measurements for various applications. So, with this we conclude the topic of thermal infrared remote sensing. So, in this particular topic we have seen what thermal infrared remote sensing is what are all the basic theoretical concepts the different definitions of temperature emissivity black body how to retrieve land surface temperature from single channel thermal sensors thermal inertia and its importance in thermal observations and also the directional anisotropy effects. All these concepts we have covered in the topic of thermal infrared remote sensing. With this we conclude this lecture and also this topic. Thank you very much.