GRMI- Flood Vulnerability Mapping Arcmap session (2021) Part1

[Music] so today um i'll be repeating the same process like i said the other thing and for today's analysis i'm using a local government in born and in benue state i know this requirement is logo so our first show was the local government to see for us to see our australia so this is our story extent like i said it's one of the governments in benue state so first thing first uh i'll be talking about i'll bring in uh the dm data that you used for this analysis the dml that beautiful analysis is the srtmdm for nigeria if you can see that you can notice that um this folder i have so i'm adding this data to my arcmap stream okay so i'll just drag this above this so that you can see uh a story area and which covers on to nigeria and mind you i i always suggest that before you make use of any data check the source to understand the resolution of data and also to understand the relationship related and also dates so that you can use it you know use how your analysis can be more clear when you are doing that so for this uh um analysis for this exercise the dm i'm using is a nice similarization with about the srt so it's quite close um so what you are going to be doing is you are going to be clipping out the extent that we'll be using for our analysis so how do we clip these off there are several ways to creep these out but for this analysis i'm going to be using extract by mask technique extract by mask technique so um for on your arcmap if you if you're not familiar with the actual you can always use the search option so how do you use the saturation you go to this uh this search ctrl f once you click on this it brings out the search pane so let me just close this so i can see that can you see yes so yes i'm going to be using my extract by mask on a special list so i click on this so um our implementation on ninja dm why um so this patient lives okay perfect so we want to impute our logo lg as our impute future mask the exact boundary of the data that we needed for our analysis it masks the data using the the exact polygon boundary that of our area of interest so every single cell within the uh area of interest is recorded so i'm going to save it as uh my logo dem so with why that's boots okay beautiful now we have our dm so what i will do is i'll just uncheck this so that i can see my dm and i zoom to what i zoom to layer to be able to see the entire extent of arcade camera see it yes we can see okay beautiful so from now from here uh this d and we're going to extract our elevation data of course the dm is also the elevation so how do we show the elevation data from the dm what we're just going to do is we right click on our data then we go to properties so from properties i would go to our symbology so it is the symbology that we are going to use to classify our data based on uh the elevation values that we have so i'm going to choose classify and uh since we know what desired class i want is 5 so i'm going to uh uh change the color i've been searching that i'll be able to to to give it a more to make it more distinct so i choose a color of uh that which shows the image of color which is green to red so i apply it so we can see so from this elevation that we can see that areas are like that represents low elevation as compared to this value while areas highlighted in red represents higher elevation so from this and dm now we're not going to extract our slope also rather using the actual box you can also make use of this search tool to uh saturate your slope i want us to be very careful now because the slope we'll be using i want to take note of the slope ah will be selected because there are there are lots and lots of them here so for this analysis i'm going to be choosing to go under the spatial analysis to toolbox slope and specialist so i'm going to select that now so uh-huh so now our rasta is not going to build our um the local dna i've just extracted here and the absolute star will specify the um their father to a thousand of in a degree i want to be that we all follow this out yes we are following okay now uh it gave me my slope in nine or 10 different classes both for me to find out let me go to the symbology okay nine different classes beautiful so we're going to reclassify this classify this radar into five different classes i want those to take note of our classification technique which is natural break that's what we're using for all of our analyses there are other classical techniques but for this particular analysis we are using our natural brain so as i've shown you the labels and also the color variation so i'm going to click on apply and let's see the results so as you can see you can see the result i'm going to uncheck this we can see the result now and the green one shows uh area of low elevation while the red shows area of eye elevation uh i want to quickly show something now so that we this can also help our so i'm going to understand i always want to so that we can see um the elevation uh effects can we all see right now yes we can see the elevation effects okay okay that's beautiful so i just wanted to i react to the point that [Music] [Music] okay thank you so i'm going to check this now and um check this back so the display i think about the transparency so that you can have it about beautiful beautiful so uh the next thing is our lunch is land cover so for this particular analysis i made use of our sentinel imagery there's no two image companies um story area so i'm going to um this is the cleat um sentinel imagery this is a clip sentiment image and as mr joe has already thought of that and i believe you can all should be able to download sensor imagery so i don't go back to that again uh so this is the crypt sensor imagery and also the clipping technique i used is still the same thing with that of gm um so so we can also zoom in to so i can see the difference i'm sorry this is quite eye so i will see how clear this is yes also uh something to notice this for the sentinel imagery the uh is quite different from the landsat both of them do both after exactly ascetic bands um i for the sake of our time and i've shown those uh the color composition but i want those to just take notes of this for the first color completely sentinel bands for not for lancers you are familiar that span 2 is the buy is the blue band band 3 is the green band band 4 is red brown band and band 5 is the near infrared brand i know for also create the first color composition of the near fire branch the red band and the green band while that is uh correct for lanterns for sentinel is quite different for sentinel band 2 is the blue band band 3 is the green band band 4 is red band both band instead of band 5 being um the near five no in this game is different but eight is the name infrared brand band eight of the sentinel two beta we can corroborate this information if you check on google we will also see the list of bands but take notes in one of your first column composites using set 2 imagery it's going to be a combination of band 8 band 4 and band 3 which is the nearby red and rebounds respectively did you all get that okay band 8 band 4 and band 3 and band 3 yes which is the knee infrared red and green bands respectively okay what's up um so for the uh what land is not covered your image is of the maximum likelihood classification and for me to do that uh i have my classification uh tabby already if you want to leave your distraction tab using your act my way i'm just going to do this once right click on on your on your standard pin anywhere on this page just right click your mouse then you see some image classification from here you just click on it you check it i will check it so i check it back again it appears on this place so uh you are going to also note that for the classification you select the clipped composite bands the composite banana you're going to select so we create what we call training samples how do you create trading samples we make use of this polygon so we drop okay now from our from our uh uh image you can you can well as well agree with me that this is a water body a very beautiful this is a riverberry cutting across uh this local government's river bandwidth so uh all that i want to do i'll select a polygon covering this area and once i've done i'm not selecting my polygon i write i double click on it once i'm done with your selection i'll go to this place the training sample monitor if you want to self-manage is where you see your selected polygons so once i click on it i'll use i'll see the only one i selected earlier this is where i choose right here and once i click on it are you following yes we're following you okay okay so you can select as many as 20 30 20 samples so that gives your analysis analysis more accuracy so for uh the water body you you and also ensure that i cover my entire entire dream right don't just focus your uh let me just select the training samples from one particular point because sometimes reflectance various for example you notice that the other water bodies in this image let me zoom into basically i'm sorry i can see what i'm seeing this particular they're particularly gonna float some cuts into this place [Music] this is also a river a river channel so while you want selection like you've noticed that the other red bunker red are pixels here and throw that we did the pictures of the black pieces so that you don't select a wrong forgiveness you'll select a wrong pixel value and give the computer uh the oscar gives you different results so once you don't select about whatever is you go um to another class i want us to take note of something can we all see all these red red dots these variables are individual three canopies this is to show us the power of uh or the the uh resolution of the special resolution of the certain imagery for each uh band it is 10 by 10 meter resolution as compared to landsat is 30 by 30 meters so you can see that this gives our uh this as a higher special resolution and this allowed allows us to be able to identify individual canopies distributed across the story area because these are illegitimate canopies so for uh also selective vacation and i believe that uh also because of the region we are dealing with if it has to be the southwest where andrews are closely spaced we can be able to deduce like okay this is our thick forest or um lightspeed so i'm going to select some bands here select them randomly across my entire area so um you can you select as many bands as possible you understand as many bands as possible the uh also um i believe they are familiar with uh the reflectance of vegetation using the first color composite it is always it is red it is red the brighter the red the brighter the color the more led the vegetation is when a vegetation is very very bright when the colorado is very bright it shows that the curve content is high and this can be seen in regions where water caught across a particular area for example if we zoom into this particular portion of the central imagery you would see that the vacation here appears brighter than g than this one i believe i'll follow you the imagination appears brighter than this one's this is to tell us that you know vegetation across a river channel absorbs more water and you know this reflects in the color and pigmentation of their leaves okay so um if you pick as many trials try to pick as many points as possible as many are polygons as possible for analysis so um i'm going to be going to the built up now also i forgot to mention uh for this particular analysis our classes will be three we are going to be classifying even three different classes and also the butter so this particular portion of the image you can see you can see i don't if you can this is quite yes yes yes yes network yes okay so beautiful so um using one of the first color composites most cases areas with um youtube appear as bright blue the appearance right now they reflect under the following first color complex combination so what i'm going to do is this area right i'm not going to select it you know and also take note of something i'm gonna tell you something once you're done classifying a particular class ensure that you rename it so that you don't mistake it for another one so uh what our body i already took a class there so for my for my um representation i took three classes so i'm just going to click on one bit use my shift button on my laptop and select the other ones so once i'm selecting the other one i'll click on this one this one goes to match the training samples into one class once i'm done writing the 20 samples i'm going to rename it to my vegetation so ahead the same goes for uh the land and the blue top to you take as many samples as possible you go across the entire country you pick your samples do not speak your samples you know particular location because reflectance like i said earlier views from each other so you select training training somebody as well and show that the area are picking is the bright blue so that you don't have a mix of of uh classes so um for the sake of this analysis i won't be picking much because of our time so just a few and i don't know so i'll classify this one stroke via their land because of a limited number of classes not to ensure that all superior areas that are also classified as vital so once you're done doing that you're going to do what creates a signature file is this great is this signature not going to pass into our maximum likelihood algorithm that we want you to classify this imagery based on the training samples you've created so what's a great um these you asked what we had to say them and kind of uh data so i'm just going to look group um google training training samples so and the extension for the signature file is gsg dot g is just bouncing this is the one i created for the edit one from analysis you also want to download okay once you save a hand you close this so we close that and we go to this classification under the classification click on the drop down arrow and we select our maximum likelihood classification so for maximum likelihood classification already this is really where uh the commodore knows the band we created the the bands used economics we directly should have completed that for us here but we want to add it we can either add it to our folder or we can use this plus button to add this from our table of contents so once you click on this button okay yeah once you click on this drop down you select um the color too clearly the color composites so and we've added it can we see it so i'll just remove them using this to select so our epc ninja fight remember i told you that the fb signature file is until you open it and you you put it back in all other values will be left are deformed the conflict at the rejection fraction prior probability so and also define the classified raster so once you run the analysis and you selected your data very well this is going to be this should be your result can we all see this also once you're done using can you see our mice classified imagery so once you don't uh create your signature file well this will clearly change sounds so it does not reflect on limitless so you can see very well that this area is what's new let me show us our uh attributes the color code so this area is our main reality like i said earlier so from our attribute [Music] since you've been in the identifiers all this we just go to do it click on our table then add alex add another field [Music] agree else's type text so first be able to add our our um our information we will use editor start editing and we select the land use land cover that we just created one is one is my rivalry from my language six is my youtube while watching this is my meditation so once you don't you save edits i stop stop editing so that's that's about that so now we've been able to uh get our slope our elevation and our launch is uncovered i believe y'all we are following and we all go to this particular point so the next thing is our rainfall data so for this particular analysis i i used i created um uh hypotensical area four points covering uh the entire story area so i'm only digitize this real four point to ensure that the points are equally sparsely distributed across my story area so for uh before we do our interpolation i want us to just check the address this is the for measurements in millimeters millimeters so you can see how many points there are ten points covering the industrial area from zero to nine ten points and this idea really for our values in millimeters for this analysis and make use of the uh interpolation technique under the just statistical analysis tools how do i navigate through statistical analysis do you click on uh this action box here this is the actual once you click on this into open and once it opens you go through your geostatistical analysis tools and you click on this uh plus icon and you do you see interpolation toolbox here you click on the position toolbox then you see the idw inverted distance waiting is the integration technique so what i do is i just open this inverse transmitting and now what i'll improve my features my pictures here is the rainfall points while the z values is going to be what the armor info it's the field that holds the magnitude of the measurement of uh the points are calculating so my output this uh how to do statistically i will leave it like that my uh my outer layer is d ah remember for distribution i'll i'll save it as ring for distribution then i'll leave everything as as it is but mind you because of my inverse starts with it if you run this analysis without performing this next technique you should give your data in a kind of rectangular format and i want to ensure that i'm going to getting false within my story about it so how do i go about this i'll go to my environment so under my environment i'm going to change the processing extent this processing extent to be same as my local lga so that the processing state is going to be confined with the boundaries of my logo lg after i'm done with that then i would uh go down also to this last analysis analysis i'm going to do a masking technique so the data is marked within my logo lg remember i need two analysis ah progress and extent and also my last analysis to mask data within my story boundary so i click on okay i've seen i've done that and i click ok and wait for the output as you can see very little um so we can see the results hello i believe we can all see this so do you have constitutional considerations i bring uh uh this is my four points and i don't overlay it i really four points okay so i also drag it all above the uh this so these are even four points like i said so for analysis i'm going to request this properties symbology five classes click on classify also we can alter these labels don't show um values let's show the uh value without changing the decimal place you know i'll just click on right command libre right let's click then change the laundry to how many significant figures are for you can see it this is app i click on apply i'm beautiful this is it there are five this shows this area has higher reform values than this religion so now we've been able to do our smoke elevation um rainforester and also the landscape next thing is our river body so for this analysis we are going to be using two different uh two different analysis the first is our big river big river as shown in the past climate she landed at the accessibility i showed earlier so i'm going to uncheck this changes so this is our large river covering the story here so after that then i'm not going to show us if another one this is for this smaller remover so for the smaller reverse segments i used the ideological analysis tools to delegate my stream my how did i go about this i'm going to show you this in the next few minutes so what i did is since i have my animation data i'll just bring in my local team that we did earlier nobody and i read earlier so and this is what i'm going to use it for my analysis so i'm going to go to my actual box under my board i'm going to go to special analyst so i'm going to go to i don't know the two box i don't know the two box so this is where i'm going to use to calculate the other uh process so first thing first i'm going to calculate the i'm going to do the fuel analysis what does different analysis do different analysis helps us to correct for sync errors that are present in our dna our data elevation model so my i will put my something outside here then my output i'm just going to uh leave it as it's clear a lot of our i want us to also connect using the same link so we cannot run documenting so we wait for our data to know and once you do the analysis this is our field data

Channel: Geohazards Risk Mapping Initiative

Views: 515

Rating: 5 out of 5

Keywords: gis, mapping, esri, arcmap, flood mapping, flood vulberability mapping, grmi

Id: Ju9cH6o4zmA

Length: 36min 49sec (2209 seconds)

Published: Mon Feb 15 2021