Explore the world of spatial analysis and cartography with geographic information systems (GIS). In this class you will learn the basics of the industry’s leading software tool, ArcGIS, during four week-long modules: Week 1: Learn how GIS grew from paper maps to the globally integrated electronic software packages of today. You will install ArcGIS on your computer and learn how to use online help to answer technical questions. Week 2: Open up ArcGIS and explore data using ArcMap. Learn the foundational concepts of GIS, how to analyze data, and make your first map. Week 3: Make your own maps! Symbolize data and create an eye-catching final product. Week 4: Share your data and maps and learn to store and organize your data. Take Fundamentals of GIS as a standalone course or as part of the Geographic Information Systems (GIS) Specialization. By completing the first class in the Specialization you will gain the skills needed to succeed in the full program. Students who need an ArcGIS license will receive a non-commercial, 1 year student license for participation in this course and specialization.
Assessing Spatial Relationships with the Spatial Join Tool
Loading...
來自 University of California, Davis 的課程
Fundamentals of GIS
2328 個評分
Explore Related Videos
At Coursera, you will find the best lectures in the world. Here are some of our personalized recommendations for you
University of California, Davis
2328 個評分
課程 1（共 5 門，Specialization Geographic Information Systems (GIS)）
從本節課中
ArcGIS Basics
In this module, we will explore GIS data using ArcMap and will explore and change properties of GIS layers to change map displays. We will subset data using selections, and explore feature attributes. Finally, we will learn about projections and use that knowledge as we run geoprocessing tools.
與講師見面
Nick SantosGeospatial Applications Researcher
Center for Watershed Sciences
[MUSIC]
Welcome back, everyone.
In this lesson, we're going to discuss my favorite tool in GIS,
the Spatial Join tool, and I'm going to show you how to use it
to attach data from one data layer to another by its location in space.
The Spatial Join tool is the proverbial Swiss Army knife of GIS.
I feel like you can make it do anything that you need it to do and that any sort
of question or tool that you might use, any of the many capable tools in our
toolbox for, you could probably make into a problem that Spatial Join can solve.
I think this so much that my coworkers kind of joke that I'm the Spatial Join
guy because whenever somebody comes into our office and asks for help with a GIS
problem, I'll say, well yeah you know you could probably do that with
a Spatial Join, and they're often times coming up with a more appropriate tool.
If you have one geoprocessing tool that you remember and
know how to use, I think Spatial Join should be it because it's so
capable of helping you in all sorts of situations.
To understand how a Spatial Join works,
I think it helps to understand what a standard Table Join is.
And a standard Table Join takes two data tables and brings them together and
attaches the attributes of one to another when they share a value in a common field.
So maybe they both have the same field but one has it in reference to the other table
so when you bring them together it says, well, do they both have the value five
in this field in one table, in this field in the other table?
Yes, okay then, attach all of the attributes in the record for
the second table to that record in the first table.
If that explanation doesn't make sense for you, just note that we are taking
data from one data table in specifically records from one data table and
appending it to the record from another data table based on some relationship.
And it can be an attribute relationship or
it can be a spatial relationship, like our Spatial Join tool.
Joins in general are incredibly powerful and
usually require some sort of sophisticated database management system to run them,
but Spatial Joins in particular are a really unique feature of as GIS,
a spatially aware database system of some sort.
I want to just give you one more comparison to help you understand what
we're going to do with the Spatial Join tool.
Remember we did this by location tool and we selected
features where they shared some sort of relationship to another set of features?
Well, we're using the same document here.
And what we're going to do with Spatial Join is basically the same thing.
Except for instead of selecting the features, we're going to take the features
that have that relationship and append attributes from one to the other.
This is really useful if we want to continue the analysis with some additional
data other than just the feature information.
Okay, so let's do a Spatial Join.
In arc toolbox, we can find it under analysis tools and overlay,
because we're basically overlaying one data layer on top of another and
doing something with that information.
And it's right here, Spatial Join.
And it's a geoprocessing tool just like we learned about.
And I'll collapse our toolbox again.
And what I want to do now is I'm going to spatial join
the counties information to the watershed information.
I'm going to attach the attribute table information from the counties
to the watersheds so that we will know what county each watershed is in.
This sounds like kind of a simple problem,
but let's take a look really quickly down here.
What happens when we have watersheds that cross the line between these counties?
It's non-trivial to say that it's in a particular county.
Well, Spatial Join accounts for that and helps us solve this problem.
So we have our Join Features and our Target Features.
Target Features receive the attributes and
Join Features give the attributes to the Target Features.
If I forgot that I can click on each field here and see in the Help.
The important thing here is that we're left with the geometry and
initial attributes of these Target Features, and they take on, also,
the attributes from the Join Features.
So what I want to do here is I want to transfer the attributes from counties
to the watersheds, so I'm going to select subwatersheds as the Target Features.
And I will select the counties as the Join Features.
And Arc GIS tries to helpfully select an output feature class for
me based on the default geo database for this document.
Instead, I'm going to put it in a different spot here, and
we'll do Spacial_Join_Watersheds_Counties.
And down here, it offers two types of join operations.
Join one to one and join one to many.
That gets at that problem I just told you about of what happens when
a watershed is in multiple counties?
How does it deal with that problem?
In join one to one, again if I forgot about, I can read about it over here.
But in join one to one it says, ok, give me one result, keep just one feature for
the watershed but I can define a merge rule about how to deal with
attributes from these multiple counties that the watershed is in.
If I did join one to many, the problem is a little simpler.
Basically what it does,is it just duplicates the watershed record and
creates a record with all the watershed's attributes, and then the one county's
attributes and then another one that has all the watershed's records, and
then whatever the other county's records are and so on until it has duplicates
of the watershed to count for all the different counties that it might be in.
In this case,
we're going to do join one to one because I want to show you some merge rules.
This next option down here, keep all target features,
is really useful in a situation like this where I have a selection,
because I'm just working with a couple counties.
And if we see, there are a lot of other counties around it and
the watershed is extended to all those counties, but
if I did keep all Target Features it would give me an output that is
all of the watersheds in the state, because that's the whole layer.
And then there would be county attributes attached only for
the counties in the selection here,
because those are the only ones active in the Join Features right now.
And every other subwatershed would just have a null value for
all the accounting information so what I might want to do, and
what I'll do in this case is I'm going to uncheck this box, and
what it's going to do for me, is it's only going to give me
an output of subwatersheds that actually successfully joined Tucson county.
So I'm going to kind of sub set my watersheds at the same time here.
And next on down is a really powerful feature that I often ignore
when I'm doing my own work, but it's a key part of using the JOIN_ONE_TO_ONE option.
And it's the field mapping or how we set all the merge rules up for
when we're trying to deal with those conflicts.
And I can see that it has all the attributes from the subwatershed table and
it has all all the attributes from the county table at the end.
This is kind of what we expect the output schema to look like.
Now, in reality if we are trying to join all of the counties' information
to the subwatersheds, we'd deal with each and every one of these but for now,
let's pretend we're only interested in getting accurate information on the names.
So I'm going to expand this and and it just has counties not named here, but
if I right click, I can set a merge rule directly right here and so
it could either use the first county it joins to or the last county it joins to,
or we can just get the count of the number of counties it's in but in this case we'll
use join which is sort of a concatenation, it appends each county to each other.
But I'm not going to set it here because I need to do something more advanced with it
in properties.
So I need to do two things here.
Number one is I'll set it to join so that if subwatershed is in multiple counties,
it gives me all of the names of the counties in one field.
And then I'll set a delimiter so that instead of just kind of putting it as
counties back to back, it'll give me them in a comma and space separated list.
The other thing I need to do is change the output field length,
because otherwise it's going to run out of space to put all these counties together,
and it'll just fail.
So I'll set that to something really large right now,
because I don't know how long they'll need to be in this case.
So I'll set it to 1,000 and that will be plenty of space.
And what I might also want to do while I'm here is change the output name because
once I'm joined I might get confused about whether or
not the name field is for the subwatersheds or for the county.
So I'll change it to County, County_NAME.
And I'll change the alias to County_Name too.
And I'll click okay.
And this is basically re-writing my data table here based upon all this input, and
that's really handy to be able to handle all these merge rules.
What's really cool is you can also do it with quantitative fields, so
In merging a field that has numbers,
I get some other merge rule, so I can get it to just sum them all up, or
give me a mean or standard deviation of those fields.
So I got some really powerful options for dealing with multiple counties per item.
Last up down here is something you are going to recognize.
It is the match item, and this looks a lot like the select by location match option,
except they are just written differently.
But the nice thing is that even though they are written differently,
if I look at the help over here, it describes exactly what they all do.
And it's pretty much the same list and you have a lot of different options for
specifying the relationship between these features in order for
them to match up to each other.
I'm just going to leave it as intersect for now so I'll click OK now and
run the tool.
And it's running down in the bottom here and
I can check on it in results if I want to.
And Spatial Join is also surprisingly fast for the amount of work it's doing.
So it's already done, and so I can collapse it and
it added a whole new layer to my map document for me.
So let's zoom to that layer.
So let's just zoom down into this corner and take a look at some spots that
would have multiple counties per watershed here, as in this one.
And first note that it left out the features that didn't get it joined
successfully, just like we expected, everything out here, and then also
even this one kind of tucked in between these other watersheds, it was left out.
But then let's take a look at one that that wasn't in multiple counties and
just identify it and see what happened to the attribute table.
So, if we scroll down, we can see all of the watershed information, and then also,
all of the county information, including our rewritten name here, county name.
And, it says Los Angeles county, which is what I'd expect.
But then, what happens if we take a look at this other watershed right here.
Let me just scroll over to make it easier to see.
If we look at this one here which is in multiple counties, I'll click on it and
I can see again how it has all the attributes of the watershed and
all the attributes of the county and
then county name and it says San Bernardino comma space, Los Angeles.
And that's because it joined the two together in a different way than our
Spatial Join but it joined the values together and it
put in the one county name and then when we put that delimiter of common space,
it then put that between any multiple values to attach to this.
This kind of functionality is really useful because
it allows us to bring all sorts of spatial information together
based upon their relationships to each other, and
then rewrite our attribute tables in ways that make sense for the project at hand.
So that's it for this demonstration of Spatial Joins.
I would really love to spend all day showing you all the cool stuff you can
do with a Spatial Join, but I think if you think about it,
you'll find you can apply it to almost any project you have.
And in fact, what we're going to do is we're going to put in a discussion
question right after this video and it's optional, but its a spot for
you to think about and read about people's potential uses for Spatial Joins.
So I hope that you'll participate there and
think about some potential ways you could use it in your own work flows.
In the meantime, I would like to give you some ideas for
some other ways you could use Spatial Join besides what we just did here.
So, one thing you could do, is if you had some population level data that's at some
really small unit like from the census, and it's stored on these small polygons,
but you wanted to aggregate it up to the county level,
you could do that with a Spatial Join, and have the summarization options for
a field add up all the population information into the county.
You can also take a rivers layer and a roads layer, and
figure out which roads cross a river, and
get that Information of the roads crossing the rivers as an attribute on the rivers.
And then a much more common use would be if I had some parcel information and
I wanted to attach information to it about different utility districts,
water districts, what election districts it's apart of, and
what counties it's in, I can just do a bunch of Spatial Joins to attach that
information from these other feature layers back down to those parcels.
So those are just a couple possibilities for
what you could do with the Spatial Join.
As I said, it's the Swiss Army knife, it's super versatile.
There's just so much that you could do with a Spatial Join.
So just a quick recap of what we learned in this lesson.
We learned how to use the Spatial Join tool to attach
information from one attribute table to another
based on the spatial relationship of the features involved.
And as I said, go take a look a that discussion question, I'll see you there.
