WEBVTT 00:00:00.000 --> 00:00:02.425 align:middle line:84% [SQUEAKING] [RUSTLING] [CLICKING] 00:00:02.425 --> 00:00:08.635 align:middle line:90% 00:00:08.635 --> 00:00:10.510 align:middle line:84% HELENA VAILLICROSA: In this part of the code, 00:00:10.510 --> 00:00:14.260 align:middle line:84% I'm going to guide you through the gap-filling process based 00:00:14.260 --> 00:00:15.470 align:middle line:90% on maps. 00:00:15.470 --> 00:00:19.970 align:middle line:84% So this is a process that I use quite a lot in my research, 00:00:19.970 --> 00:00:23.470 align:middle line:84% which is when I'm facing a new database sometimes 00:00:23.470 --> 00:00:27.850 align:middle line:84% comes with some empty slots, some N/As in the data. 00:00:27.850 --> 00:00:32.680 align:middle line:84% And I want to fill them based on field data. 00:00:32.680 --> 00:00:36.970 align:middle line:84% Obviously, data coming from maps is not 00:00:36.970 --> 00:00:40.340 align:middle line:84% as accurate as data coming from field experiments. 00:00:40.340 --> 00:00:44.260 align:middle line:84% But this is a good proxy we have to maximize the data we have 00:00:44.260 --> 00:00:47.840 align:middle line:90% and use them all in our models. 00:00:47.840 --> 00:00:49.960 align:middle line:84% So what I'm going to be doing here 00:00:49.960 --> 00:00:53.440 align:middle line:84% is to use the latitude and longitude of the coordinates 00:00:53.440 --> 00:00:57.190 align:middle line:84% that comes into the databases, transform them 00:00:57.190 --> 00:00:59.740 align:middle line:84% into spatial-explicit data points, 00:00:59.740 --> 00:01:02.950 align:middle line:84% and use these data points to extract 00:01:02.950 --> 00:01:07.190 align:middle line:90% the data coming from a map. 00:01:07.190 --> 00:01:08.380 align:middle line:90% So let's just go ahead. 00:01:08.380 --> 00:01:11.600 align:middle line:84% I'm going to download real data that I created 00:01:11.600 --> 00:01:13.310 align:middle line:90% that is in my GitHub account. 00:01:13.310 --> 00:01:15.980 align:middle line:90% You can download it freely here. 00:01:15.980 --> 00:01:19.760 align:middle line:84% It's a database that calculates nitrogen uptake global scale 00:01:19.760 --> 00:01:22.790 align:middle line:90% based on field data. 00:01:22.790 --> 00:01:25.670 align:middle line:84% But I encountered that some of the drivers-- 00:01:25.670 --> 00:01:28.190 align:middle line:84% temperature, precipitation, soil information-- 00:01:28.190 --> 00:01:30.770 align:middle line:84% has not always been collected, or maybe 00:01:30.770 --> 00:01:34.080 align:middle line:84% wasn't reported in the original papers. 00:01:34.080 --> 00:01:38.370 align:middle line:84% So that's why there are some data gaps. 00:01:38.370 --> 00:01:42.980 align:middle line:84% So well, you can see that it's is a big database. 00:01:42.980 --> 00:01:44.640 align:middle line:90% It includes a lot of columns. 00:01:44.640 --> 00:01:46.280 align:middle line:90% Some of them are complete. 00:01:46.280 --> 00:01:47.510 align:middle line:90% Some of them have N/As. 00:01:47.510 --> 00:01:51.180 align:middle line:90% 00:01:51.180 --> 00:01:56.330 align:middle line:84% So I'm just going to see how are they called. 00:01:56.330 --> 00:01:58.710 align:middle line:84% These are all the names of the columns. 00:01:58.710 --> 00:02:00.410 align:middle line:84% And we are going to be particularly 00:02:00.410 --> 00:02:02.570 align:middle line:84% interested in the latitude and longitude, which 00:02:02.570 --> 00:02:04.550 align:middle line:84% is the third and fourth column right here. 00:02:04.550 --> 00:02:08.020 align:middle line:90% 00:02:08.020 --> 00:02:11.640 align:middle line:84% I'm going to make sure that we don't have N/As there. 00:02:11.640 --> 00:02:15.920 align:middle line:84% So I'm just going to subset the data that doesn't include N/As. 00:02:15.920 --> 00:02:22.220 align:middle line:84% So it is N/A. So it doesn't include N/A in the latitude. 00:02:22.220 --> 00:02:25.200 align:middle line:84% And it doesn't include N/As the longitude. 00:02:25.200 --> 00:02:27.320 align:middle line:84% This is one of the several ways we 00:02:27.320 --> 00:02:29.930 align:middle line:84% have to make sure that we exclude N/As 00:02:29.930 --> 00:02:33.740 align:middle line:90% in these particular columns. 00:02:33.740 --> 00:02:34.460 align:middle line:90% OK. 00:02:34.460 --> 00:02:36.320 align:middle line:84% Now that we have this sorted out, 00:02:36.320 --> 00:02:39.900 align:middle line:84% there are two ways of transforming these data points, 00:02:39.900 --> 00:02:43.290 align:middle line:84% so this latitude and longitude in data-explicit points. 00:02:43.290 --> 00:02:45.350 align:middle line:90% One is using SpatialPoints here. 00:02:45.350 --> 00:02:49.250 align:middle line:84% And the other one is using SpatialPointsDataFrame. 00:02:49.250 --> 00:02:51.800 align:middle line:84% In both of them, I need to specify 00:02:51.800 --> 00:02:55.490 align:middle line:84% where latitude and longitude are located, what's 00:02:55.490 --> 00:02:57.150 align:middle line:90% the database that I'm using. 00:02:57.150 --> 00:03:01.100 align:middle line:84% And in the first case, I need to specify also the resolution 00:03:01.100 --> 00:03:02.720 align:middle line:90% I want the data to be. 00:03:02.720 --> 00:03:04.640 align:middle line:90% I personally like this more. 00:03:04.640 --> 00:03:06.290 align:middle line:84% But with this one right here, it's 00:03:06.290 --> 00:03:10.720 align:middle line:84% going to also work for our purposes right now. 00:03:10.720 --> 00:03:15.760 align:middle line:84% We see that the outcome we get is a slightly different one 00:03:15.760 --> 00:03:21.630 align:middle line:84% in SpatialPointsDataFrame and the other one is SpatialPoints. 00:03:21.630 --> 00:03:25.310 align:middle line:84% But if we plot them they look very similar. 00:03:25.310 --> 00:03:30.760 align:middle line:90% 00:03:30.760 --> 00:03:37.195 align:middle line:84% And we can plot them with the shape of the Earth. 00:03:37.195 --> 00:03:38.920 align:middle line:84% We don't have a good understanding 00:03:38.920 --> 00:03:40.855 align:middle line:90% of what is this plotting. 00:03:40.855 --> 00:03:44.200 align:middle line:90% 00:03:44.200 --> 00:03:47.410 align:middle line:84% Here, I added this map that we already used 00:03:47.410 --> 00:03:51.620 align:middle line:90% in other parts of the videos. 00:03:51.620 --> 00:03:54.940 align:middle line:84% So here, we can have a good understanding of where 00:03:54.940 --> 00:03:57.490 align:middle line:90% this data is coming from. 00:03:57.490 --> 00:03:59.380 align:middle line:84% And it also certifies that we have 00:03:59.380 --> 00:04:01.520 align:middle line:90% built our map appropriately. 00:04:01.520 --> 00:04:04.310 align:middle line:84% Sometimes, if you mismatch latitude and longitude, 00:04:04.310 --> 00:04:09.020 align:middle line:84% you're going to have a twisted map. 00:04:09.020 --> 00:04:11.710 align:middle line:90% So let's just try it for fun. 00:04:11.710 --> 00:04:25.003 align:middle line:90% 00:04:25.003 --> 00:04:26.170 align:middle line:90% Let's see how it looks like. 00:04:26.170 --> 00:04:29.220 align:middle line:90% 00:04:29.220 --> 00:04:34.080 align:middle line:84% You see, now, it's totally changed 90 degrees. 00:04:34.080 --> 00:04:37.230 align:middle line:84% But just a heads up to put latitude and longitude 00:04:37.230 --> 00:04:42.720 align:middle line:90% in the appropriate location. 00:04:42.720 --> 00:04:43.830 align:middle line:90% OK. 00:04:43.830 --> 00:04:48.120 align:middle line:84% So going back to the process, I'm 00:04:48.120 --> 00:04:51.060 align:middle line:84% going to use as a base this map that we've 00:04:51.060 --> 00:04:54.660 align:middle line:84% been using recurrently, which is this temperature map. 00:04:54.660 --> 00:04:58.990 align:middle line:84% And I want to extract the information using these points. 00:04:58.990 --> 00:05:02.220 align:middle line:84% So I only need to use the function extract, the map where 00:05:02.220 --> 00:05:03.750 align:middle line:84% I want to take the information, from 00:05:03.750 --> 00:05:07.930 align:middle line:84% and the points I want to use to extract the data. 00:05:07.930 --> 00:05:09.240 align:middle line:90% So I use these. 00:05:09.240 --> 00:05:17.550 align:middle line:84% And I see that the result is this right here-- oh, sorry-- 00:05:17.550 --> 00:05:21.020 align:middle line:90% 00:05:21.020 --> 00:05:21.590 align:middle line:90% here. 00:05:21.590 --> 00:05:24.990 align:middle line:84% And we see that we have a new value here, 00:05:24.990 --> 00:05:31.280 align:middle line:84% a new vector that is 159 data points long, which is exactly 00:05:31.280 --> 00:05:33.140 align:middle line:90% the same number of columns-- 00:05:33.140 --> 00:05:36.530 align:middle line:84% of rows, sorry, we have in our data frame. 00:05:36.530 --> 00:05:40.670 align:middle line:84% And if instead of having one map here 00:05:40.670 --> 00:05:44.570 align:middle line:84% we would have had a stack or a brick, 00:05:44.570 --> 00:05:47.880 align:middle line:84% we would have had a column per map. 00:05:47.880 --> 00:05:51.350 align:middle line:84% So in this case, we only have one because we have one map. 00:05:51.350 --> 00:05:53.480 align:middle line:84% But the more maps we have, the more columns 00:05:53.480 --> 00:05:59.660 align:middle line:90% it would add in our result. 00:05:59.660 --> 00:06:02.660 align:middle line:84% Now, I'm going to guide you through an exercise that 00:06:02.660 --> 00:06:05.990 align:middle line:84% hopefully is going to be a good summary of all 00:06:05.990 --> 00:06:07.640 align:middle line:84% the functions and all the concepts 00:06:07.640 --> 00:06:14.510 align:middle line:84% we've been working with in this GIS session. 00:06:14.510 --> 00:06:15.990 align:middle line:84% I'm going to guide you through it. 00:06:15.990 --> 00:06:18.150 align:middle line:84% So initially, there's nothing done here. 00:06:18.150 --> 00:06:23.090 align:middle line:84% So I'm going to type while doing it live. 00:06:23.090 --> 00:06:26.240 align:middle line:84% And first of all, I would ask you 00:06:26.240 --> 00:06:29.330 align:middle line:84% to download these 10-meter resolution average temperature 00:06:29.330 --> 00:06:31.550 align:middle line:90% maps from WorldClim. 00:06:31.550 --> 00:06:32.930 align:middle line:90% And this is the link. 00:06:32.930 --> 00:06:34.820 align:middle line:84% But in case you don't have it handy, 00:06:34.820 --> 00:06:37.760 align:middle line:84% this is how the web page looks like. 00:06:37.760 --> 00:06:40.370 align:middle line:84% I would ask you to go to this average temperature 00:06:40.370 --> 00:06:42.470 align:middle line:90% and download this one. 00:06:42.470 --> 00:06:45.050 align:middle line:84% Feel free to download other resolutions as well. 00:06:45.050 --> 00:06:47.270 align:middle line:84% But I just went to the course one 00:06:47.270 --> 00:06:51.170 align:middle line:84% to make the explanation more fluid and the calculations 00:06:51.170 --> 00:06:55.410 align:middle line:90% easier to calculate. 00:06:55.410 --> 00:06:57.590 align:middle line:84% So let's download these and store it 00:06:57.590 --> 00:07:01.020 align:middle line:84% somewhere in your computer that you know the oute of. 00:07:01.020 --> 00:07:05.640 align:middle line:84% And also, I'm going to ask you to load all these maps in your R 00:07:05.640 --> 00:07:08.970 align:middle line:84% session and calculate the average of all the maps, 00:07:08.970 --> 00:07:13.020 align:middle line:84% so to obtain the yearly average temperature. 00:07:13.020 --> 00:07:13.680 align:middle line:90% OK. 00:07:13.680 --> 00:07:16.440 align:middle line:84% To do that, I'm going to use code that we've already used. 00:07:16.440 --> 00:07:19.950 align:middle line:90% I'm just going to jump around. 00:07:19.950 --> 00:07:22.200 align:middle line:90% I think I have this code here. 00:07:22.200 --> 00:07:24.240 align:middle line:90% Yes. 00:07:24.240 --> 00:07:25.185 align:middle line:90% I'm going to use this. 00:07:25.185 --> 00:07:34.160 align:middle line:90% 00:07:34.160 --> 00:07:36.350 align:middle line:84% So I'm going to charge everything at once, 00:07:36.350 --> 00:07:38.110 align:middle line:90% instead of going one by one. 00:07:38.110 --> 00:07:40.450 align:middle line:90% I'm going to do it faster. 00:07:40.450 --> 00:07:42.790 align:middle line:90% This and this-- 00:07:42.790 --> 00:07:45.700 align:middle line:84% I don't really want this code here. 00:07:45.700 --> 00:07:48.840 align:middle line:90% 00:07:48.840 --> 00:07:51.990 align:middle line:84% I'm going to create a stack of these. 00:07:51.990 --> 00:07:56.190 align:middle line:84% And now, we want to calculate the mean of all these maps. 00:07:56.190 --> 00:07:59.145 align:middle line:84% So to do that, it's going to be called-- 00:07:59.145 --> 00:08:06.740 align:middle line:90% 00:08:06.740 --> 00:08:16.305 align:middle line:84% I'm going to use this object here and calculate the mean. 00:08:16.305 --> 00:08:19.770 align:middle line:90% 00:08:19.770 --> 00:08:20.610 align:middle line:90% This is what we got. 00:08:20.610 --> 00:08:23.180 align:middle line:90% 00:08:23.180 --> 00:08:25.940 align:middle line:84% So see that the information looks good. 00:08:25.940 --> 00:08:28.210 align:middle line:84% The values-- minimum, the maximum-- 00:08:28.210 --> 00:08:29.720 align:middle line:90% they look feasible. 00:08:29.720 --> 00:08:33.340 align:middle line:84% So it seems that everything worked out pretty well. 00:08:33.340 --> 00:08:35.860 align:middle line:84% I'm just going to plot it to see how it looks like. 00:08:35.860 --> 00:08:43.510 align:middle line:90% 00:08:43.510 --> 00:08:44.800 align:middle line:90% So yeah, it seems reasonable. 00:08:44.800 --> 00:08:48.790 align:middle line:84% So I think we've successfully completed the first step. 00:08:48.790 --> 00:08:54.030 align:middle line:84% Second step-- I want you to download the nitrogen uptake 00:08:54.030 --> 00:08:58.230 align:middle line:84% data from this page, which is my GitHub page. 00:08:58.230 --> 00:09:01.000 align:middle line:84% This is the database we've used in the previous exercise. 00:09:01.000 --> 00:09:02.790 align:middle line:90% So you might have it already. 00:09:02.790 --> 00:09:05.520 align:middle line:84% And I'm going to ask you to create a new dataframe 00:09:05.520 --> 00:09:10.200 align:middle line:84% by subsetting the columns numbered 3, 4, and 18, which 00:09:10.200 --> 00:09:13.950 align:middle line:84% are the latitude, longitude, and mean annual temperature, which 00:09:13.950 --> 00:09:18.670 align:middle line:84% is what we just calculated in our maps. 00:09:18.670 --> 00:09:25.860 align:middle line:84% So to do that, I'm just going to load the data 00:09:25.860 --> 00:09:27.375 align:middle line:90% that I should have here. 00:09:27.375 --> 00:09:35.950 align:middle line:90% 00:09:35.950 --> 00:09:38.623 align:middle line:90% This is the data. 00:09:38.623 --> 00:09:39.540 align:middle line:90% We're going to subset. 00:09:39.540 --> 00:09:52.720 align:middle line:90% 00:09:52.720 --> 00:09:58.225 align:middle line:84% I said the column number 3, number 4, and number 18. 00:09:58.225 --> 00:10:01.640 align:middle line:90% 00:10:01.640 --> 00:10:05.430 align:middle line:84% Now, we have this object that has exactly the same length-- 00:10:05.430 --> 00:10:09.500 align:middle line:84% so the same rows, but only these three variables 00:10:09.500 --> 00:10:10.640 align:middle line:90% that I asked you for-- 00:10:10.640 --> 00:10:14.720 align:middle line:84% latitude, longitude, and mean annual temperature. 00:10:14.720 --> 00:10:16.670 align:middle line:84% I chose these variables because we 00:10:16.670 --> 00:10:19.400 align:middle line:90% have some N/As, as I mentioned. 00:10:19.400 --> 00:10:22.220 align:middle line:84% You see some empty values that we would 00:10:22.220 --> 00:10:28.200 align:middle line:90% like to fill in further steps. 00:10:28.200 --> 00:10:28.700 align:middle line:90% OK. 00:10:28.700 --> 00:10:31.520 align:middle line:90% Let's move on to the step 3. 00:10:31.520 --> 00:10:34.880 align:middle line:84% I want to create spatial points by using the coordinates 00:10:34.880 --> 00:10:37.130 align:middle line:90% in the subsetted data frame. 00:10:37.130 --> 00:10:40.880 align:middle line:84% And I also want to extract the mean annual temperature 00:10:40.880 --> 00:10:44.270 align:middle line:84% of the data points you just created from the map 00:10:44.270 --> 00:10:48.590 align:middle line:84% and include them in a new vector to the data frame, 00:10:48.590 --> 00:10:51.840 align:middle line:84% resulting in a four-column data frame. 00:10:51.840 --> 00:11:02.280 align:middle line:84% So I'm going to go through the code that we just used. 00:11:02.280 --> 00:11:04.685 align:middle line:90% Oops, sorry, sorry, sorry. 00:11:04.685 --> 00:11:18.090 align:middle line:90% 00:11:18.090 --> 00:11:19.170 align:middle line:90% Find them here. 00:11:19.170 --> 00:11:22.930 align:middle line:90% 00:11:22.930 --> 00:11:24.593 align:middle line:90% This one. 00:11:24.593 --> 00:11:26.385 align:middle line:84% We're going to use this one that's shorter. 00:11:26.385 --> 00:11:36.130 align:middle line:90% 00:11:36.130 --> 00:11:36.850 align:middle line:90% OK. 00:11:36.850 --> 00:11:38.080 align:middle line:90% We created that. 00:11:38.080 --> 00:11:40.580 align:middle line:84% And now, we want to extract this information. 00:11:40.580 --> 00:11:42.550 align:middle line:90% So it's going to be called-- 00:11:42.550 --> 00:11:50.660 align:middle line:90% 00:11:50.660 --> 00:11:55.620 align:middle line:84% MAT stands for mean annual temperature. 00:11:55.620 --> 00:11:59.050 align:middle line:84% And I want to extract from this variable that we created-- 00:11:59.050 --> 00:12:00.330 align:middle line:90% that's the mean temperature-- 00:12:00.330 --> 00:12:04.980 align:middle line:90% 00:12:04.980 --> 00:12:05.980 align:middle line:90% and the PNT. 00:12:05.980 --> 00:12:10.100 align:middle line:90% 00:12:10.100 --> 00:12:11.210 align:middle line:90% OK. 00:12:11.210 --> 00:12:12.920 align:middle line:90% We created this vector here. 00:12:12.920 --> 00:12:17.180 align:middle line:84% And now, we want to add this vector to our initial data 00:12:17.180 --> 00:12:18.710 align:middle line:90% frame. 00:12:18.710 --> 00:12:21.230 align:middle line:90% So our data frame was created-- 00:12:21.230 --> 00:12:23.090 align:middle line:90% was called this. 00:12:23.090 --> 00:12:27.040 align:middle line:90% 00:12:27.040 --> 00:12:32.735 align:middle line:84% It's going to be called Coming From WorldClim. 00:12:32.735 --> 00:12:49.000 align:middle line:90% 00:12:49.000 --> 00:12:50.355 align:middle line:90% This is one way to do it. 00:12:50.355 --> 00:12:51.230 align:middle line:90% There are other ways. 00:12:51.230 --> 00:12:55.660 align:middle line:84% So if you find an easier way that works better 00:12:55.660 --> 00:12:57.830 align:middle line:90% for you to do it, feel free. 00:12:57.830 --> 00:13:00.550 align:middle line:84% So now, we have this initial database 00:13:00.550 --> 00:13:02.390 align:middle line:90% that had only three columns. 00:13:02.390 --> 00:13:03.970 align:middle line:90% Now, it has four columns. 00:13:03.970 --> 00:13:09.160 align:middle line:84% And we just added this new column here with a new variable. 00:13:09.160 --> 00:13:11.710 align:middle line:84% We could even see how these two match. 00:13:11.710 --> 00:13:15.380 align:middle line:84% So this comes from ground measurements. 00:13:15.380 --> 00:13:16.810 align:middle line:90% So it's based on field data. 00:13:16.810 --> 00:13:17.920 align:middle line:90% And this is the map. 00:13:17.920 --> 00:13:20.410 align:middle line:84% We could even see how these two match. 00:13:20.410 --> 00:13:23.650 align:middle line:84% We could, for example, plot these two informations 00:13:23.650 --> 00:13:25.780 align:middle line:90% or create a linear model. 00:13:25.780 --> 00:13:27.770 align:middle line:84% I'm just going to go with the model. 00:13:27.770 --> 00:13:29.920 align:middle line:90% It's going to be called Model 1. 00:13:29.920 --> 00:13:45.850 align:middle line:84% It's going to be a linear model, where I have these, 00:13:45.850 --> 00:13:50.500 align:middle line:90% based on WorldClim. 00:13:50.500 --> 00:13:54.655 align:middle line:84% This data is going to be this one. 00:13:54.655 --> 00:14:05.010 align:middle line:90% 00:14:05.010 --> 00:14:05.590 align:middle line:90% Let's see. 00:14:05.590 --> 00:14:07.350 align:middle line:90% Well, it's almost the same data. 00:14:07.350 --> 00:14:10.350 align:middle line:90% So they agree in a 94%. 00:14:10.350 --> 00:14:13.260 align:middle line:90% 00:14:13.260 --> 00:14:16.180 align:middle line:84% Yes, almost 1, a perfect regression. 00:14:16.180 --> 00:14:18.480 align:middle line:90% So this is a good proxy. 00:14:18.480 --> 00:14:24.960 align:middle line:84% We could also plot them in a linear model-- 00:14:24.960 --> 00:14:26.960 align:middle line:90% in a regression plot. 00:14:26.960 --> 00:14:40.080 align:middle line:90% It's just like this. 00:14:40.080 --> 00:14:44.190 align:middle line:84% And we see how the data mostly agrees. 00:14:44.190 --> 00:14:49.310 align:middle line:90% 00:14:49.310 --> 00:14:49.810 align:middle line:90% OK. 00:14:49.810 --> 00:14:53.990 align:middle line:84% And now finally, I want to combine these two columns. 00:14:53.990 --> 00:14:56.350 align:middle line:84% So create a new column in your data frame 00:14:56.350 --> 00:15:00.250 align:middle line:84% where you gap-fill the mean annual temperature 00:15:00.250 --> 00:15:02.980 align:middle line:84% variable with the data you just extracted. 00:15:02.980 --> 00:15:07.420 align:middle line:84% So what I want to do here is to use-- 00:15:07.420 --> 00:15:12.640 align:middle line:84% so mainly use this field experimental data, 00:15:12.640 --> 00:15:17.860 align:middle line:84% but use the map to gap-fill these empty spaces that I 00:15:17.860 --> 00:15:20.380 align:middle line:90% have here. 00:15:20.380 --> 00:15:24.020 align:middle line:84% To do so, I'm going to use code that I already have, 00:15:24.020 --> 00:15:41.660 align:middle line:84% which uses a for loop, so it's faster. 00:15:41.660 --> 00:15:42.160 align:middle line:90% OK. 00:15:42.160 --> 00:15:44.470 align:middle line:90% What I'm doing is to-- 00:15:44.470 --> 00:15:45.850 align:middle line:90% ah, this data-- 00:15:45.850 --> 00:15:53.240 align:middle line:84% I'm going to change the code using this tool. 00:15:53.240 --> 00:15:55.480 align:middle line:84% So what initially was General DB, now 00:15:55.480 --> 00:16:07.680 align:middle line:84% it's going to be called N in the section, which 00:16:07.680 --> 00:16:09.000 align:middle line:90% is what I selected. 00:16:09.000 --> 00:16:11.790 align:middle line:90% And I replace them all. 00:16:11.790 --> 00:16:14.580 align:middle line:90% Yes, four occurrences. 00:16:14.580 --> 00:16:15.180 align:middle line:90% OK. 00:16:15.180 --> 00:16:17.650 align:middle line:90% And now, the data-- 00:16:17.650 --> 00:16:19.920 align:middle line:84% so what I'm going to be doing is I'm 00:16:19.920 --> 00:16:25.860 align:middle line:84% asking R to generate this for loop that goes from 1 to 159, 00:16:25.860 --> 00:16:28.380 align:middle line:84% which is the number of rows that I have. 00:16:28.380 --> 00:16:31.650 align:middle line:90% And I'm saying, if this is N/A-- 00:16:31.650 --> 00:16:32.955 align:middle line:90% so it's an empty space-- 00:16:32.955 --> 00:16:39.070 align:middle line:90% 00:16:39.070 --> 00:16:46.030 align:middle line:84% if this variable is an empty space, go to the same variable 00:16:46.030 --> 00:17:03.700 align:middle line:84% and fill this space with what you will find here. 00:17:03.700 --> 00:17:07.470 align:middle line:84% And if that's not an N/A, just leave it like is. 00:17:07.470 --> 00:17:16.270 align:middle line:90% 00:17:16.270 --> 00:17:19.900 align:middle line:90% You could even erase this part. 00:17:19.900 --> 00:17:24.069 align:middle line:84% So we could not need this else part 00:17:24.069 --> 00:17:26.740 align:middle line:84% because it's basically leaving the thing as they are. 00:17:26.740 --> 00:17:29.440 align:middle line:84% So you could decide to erase this. 00:17:29.440 --> 00:17:32.780 align:middle line:84% And things would work either way. 00:17:32.780 --> 00:17:34.240 align:middle line:90% So I'm going to make this run. 00:17:34.240 --> 00:17:36.580 align:middle line:90% Maybe it gives us some error. 00:17:36.580 --> 00:17:38.170 align:middle line:90% It seemed that it went smooth. 00:17:38.170 --> 00:17:42.770 align:middle line:90% 00:17:42.770 --> 00:17:45.510 align:middle line:90% OK. 00:17:45.510 --> 00:17:47.750 align:middle line:84% Now, we see that this variable here 00:17:47.750 --> 00:17:51.980 align:middle line:84% doesn't have N/As because we filled 00:17:51.980 --> 00:17:55.200 align:middle line:84% these information with the information coming here. 00:17:55.200 --> 00:17:58.010 align:middle line:84% In this case, I have destroyed the initial variable, 00:17:58.010 --> 00:18:00.140 align:middle line:84% meaning that now we no longer have 00:18:00.140 --> 00:18:02.210 align:middle line:90% the variable with these spaces. 00:18:02.210 --> 00:18:04.700 align:middle line:84% I could have created a new variable, which 00:18:04.700 --> 00:18:08.540 align:middle line:84% is what I'm going to do now just for you to see it 00:18:08.540 --> 00:18:10.950 align:middle line:90% and to compare things better. 00:18:10.950 --> 00:18:16.080 align:middle line:84% So I'm going to rerun the whole code. 00:18:16.080 --> 00:18:17.000 align:middle line:90% A very easy way-- 00:18:17.000 --> 00:18:20.930 align:middle line:90% 00:18:20.930 --> 00:18:24.770 align:middle line:84% these are the advantages of R, that you 00:18:24.770 --> 00:18:27.800 align:middle line:90% can rerun things pretty easy. 00:18:27.800 --> 00:18:33.147 align:middle line:84% And now, instead of rearranging things 00:18:33.147 --> 00:18:34.980 align:middle line:84% on top of the variable that already existed, 00:18:34.980 --> 00:18:37.480 align:middle line:84% I'm going to create a new one that's going to be called New. 00:18:37.480 --> 00:18:42.170 align:middle line:90% 00:18:42.170 --> 00:18:46.400 align:middle line:84% New-- I'm going to redo the process. 00:18:46.400 --> 00:18:49.260 align:middle line:90% 00:18:49.260 --> 00:18:54.700 align:middle line:84% Now, we have this new code, this new column here. 00:18:54.700 --> 00:19:00.210 align:middle line:84% And so when we have an N/A, it's always the same as MAT. 00:19:00.210 --> 00:19:01.650 align:middle line:84% But in the ones that we have N/A, 00:19:01.650 --> 00:19:03.870 align:middle line:90% we've used this one instead. 00:19:03.870 --> 00:19:10.080 align:middle line:84% And this is a very efficient way that we have to gap-fill data 00:19:10.080 --> 00:19:14.060 align:middle line:90% coming from map data. 00:19:14.060 --> 00:19:19.000 align:middle line:90%