Watch videos with subtitles in your language, upload your videos, create your own subtitles! Click here to learn more on "how to Dotsub"

BITC / Biodiversity Diagnoses - Environmental Variation

0 (0 Likes / 0 Dislikes)
As Town mentioned, we're going to talk about measuring environmental variability across study regions. We'll just start off talking about biomes a little bit. At a very broad scale, people generally classify environmental variation into different biomes, which are driven mainly by temperature and rainfall, which determine which will be the dominant vegetation in an area. So, if we are looking at Africa, we can see that there is desert and savannah and rainforest, but if you are going to be doing a regional analysis, these categories are far too broad to capture any variation across your area. If you zoom in a little bit, the next level are ecosystems this is just an example of different ecosystems across Africa, according to the World Wildlife Fund. So you can see, if you were zooming in to this desert region, then there's quite a few different ecosystems going on in this area. If you zoom in further, to Uganda, as Town was mentioning, then there is quite a bit of environmental variation going on, in just a small area, so there's rain forest, montane areas, savannah, just a wide variety of different environments available in this area. So, what we are going to talk about is different ways to measure these environmental variables, and examples of data that have been measured across different regions. Then how to observe the environmental data in your region, and how to compare this with your occurrence points, and the environmental data across the entire study area. So, there are many ways to measure environmental variables, but recently, over the past 30 or 40 years, a lot of people have been utilizing remote sensing data, which is captured from satellites that fly over the Earth, and collect data on reflectance from the Earth's surface and they collect these different numbers, and calibrate them into these different data products, like elevation or measurements of primary productivity, or greenness of vegetation in an area. You can also measure environmental variables by interpolating your data across an area, and this involves having for example data from a lot of weather stations, and estimating values between the stations. The results of these different data capturing methods are turned into a continuous gridded surface, which is called a raster grid. You can think about it as a system of quadrats This is just a continuous grid that has different values for temperature, or precipitation, across your region. So, this is an example of an interpolated surface for temperature across Uganda. You can see up here, there are higher temperatures in this region, and then where there is blue, there are lower temperatures. This is the Lake Victoria area here. These are occurrence points that we will talk about more in just a minute, but you can see where your occurrence points lie across a gridded surface and get an idea of which environments your sampling points are representing across a study region. This is another example ... of precipitation ... It's also an interpolated surface. This is a product that measures the "green up" of the vegetation in an area, so this can be very useful if you are looking at a species that finds secondary forest very important There's a particular range of these values that represent secondary forest, so you can identify the range of variation across your study region, and whether or not you're really capturing all of that secondary forest, or whether there's other, more scrublike vegetation in the region. If you're looking at fisheries, or whales, as Arturo was talking about, then you can look at different sea variables sea surface temperature is an example: the black part here is the land surface, and these different values represent the temperature And then, of course, elevation is an important variable that a lot of people look at, and you can derive a lot of variables from an elevation surface So one of these is the compound topographic index, which measures where water is likely to collect if there's a precipitation event. So these areas show, if it were to rain, here's where water would be collecting (in the blue areas), and then in the red areas, this is where it's like to run off. So if you have a species that prefers to be in moist soil, or very dry soil this data layer can tell you a lot about what is going on in your study region and whether or not you have sampled all of the moist areas or the dry areas Of course, all of these different data products come in a variety of scales and resolutions, generally, the climate data tend to be very coarse Often, when you are looking at the climate data from a remote sensing product, where the satellite went over and is collecting data, then it's going to be very coarse, so it's going to be hard to get an idea of the variation across your area If you're looking at interpolated surfaces, then you can get a lot more information from those data layers. So these [data layers] go from 1 degree resolution all the way down to 1 meter resolution, so you have to think about your species, what might be important to it, and what is the scale of variation that you might want to look at for your species, to know if you have captured the variation of interest across your study region. To observe this environmental variation, you have your occurrence points that are located at different x and y coordinates this is an example ... just envision this gridded surface, across where this red square is Each of these squares has a value. So, if these are the occurrence points for your species, you can take the value for your species at that location, and put it into a table over here. This is going to represent all of the environmental values where your species is located. The next step, if you want to get a better idea of what's going on across your study region, is that you can--even though Town just talked about not putting random points across your study region-- if you want to just visualize what might be going on in your study region, then you can put a whole bunch of random points across that same area, you would do the same thing... if these are the random points, then you would extract these environmental values at the random points, and then you put them into the table over here and then you'll be able to visualize what's going on between the environmental values at your points and at the random points and that's what we are going to be looking at a little bit today. We are going to use the example of the shoebill stork, which occurs in this region This is the range of the shoebill stork, according to the IUCN Basically, this bird likes freshwater swamps We are going to take a look at how you can use remote sensing data to observe these environmental variables across the region, all the way down to what might be important for these storks in this area, and what's going on across our entire study region. We are just going to use two variables, to make it pretty simple. We are going to look at temperature and precipitation. These are the interpolated surfaces that I was talking about before. These are occurrence points that I pulled down from GBIF and these are their locations, so you can see that they are distributed close to Lake Victoria, and also there's another cluster of points up here But if you are looking at this surface, then you can see right away that there are certain areas that haven't been sampled very well by these occurrence points. So, there's not a lot of points in these dark blue areas, which are low-temperature areas, or in these really red areas, which have very high temperatures So these are areas that if you were just looking at the environmental values of your sample points, it's not really representative of the environment across your entire study region We're also going to do the same thing for precipitation here, you can see higher precipitation in this area, there's no environmental values associated with any of the occurrence points that would occur in this higher precipitation region And then the other thing that you have to think about is that when you're looking at these environmental variables that you're interested in for your species, it's not just one variable or the other variable you have to also think about the combinations of the variables So, here, I've overlaid the temperature surface onto the precipitation surface and there's a little bit different colors, so it may be that these occurrence points do a really good job of representing temperature in an area, but they may not do as good of a job representing precipitation in the area So, when you're looking at the combination of the two variables, there may be these environments that are not found across the entire study region So it's not just one independent variable versus another independent variable, The more variables that you are interested in, the more you are going to have to think about maybe where is it warm AND rainy, or where is it cold AND rainy, and is this represented across the occurrence points So everyone can see a bit better, we're going to go ahead and do this in QGIS... and show how to have your occurrence points, extract the environmental values from your occurrence points, and then also how to extract environmental values from a bunch of random points across your study area, and then we're going to take a look at visualizing to see whether the sampled areas from your occurrence points are representative of the environments across your study region. So Town is going to do this on his laptop computer ...

Video Details

Duration: 13 minutes and 46 seconds
Language: English
License: Dotsub - Standard License
Genre: None
Views: 0
Posted by: townpeterson on Jul 26, 2016

This talk was presented in the course on National Biodiversity Diagnoses, an advanced course focused on developing summaries of state of knowledge of particular taxa for countries and regions. The workshop was held in Entebbe, Uganda, during 12-17 January 2015. Workshop organized by the Biodiversity Informatics Training Curriculum, with funding from the JRS Biodiversity Foundation.

Caption and Translate

    Sign In/Register for Dotsub above to caption this video.