GIS in Archaeology

class: title-slide, center, middle

#  GIS in Archaeology

##  05 - Handling Spatial Data

###  Martin Hinz

####  Institut für Archäologische Wissenschaften, Universität Bern

23/10/24

.footnote[
.right[
.tiny[
You can download a [pdf of this presentation](gis_in_archaeology05.pdf).
]
]
]
---
## Things to cover

+ Manipulation of vector data

+ Spatial selection of vector data

+ Working with buffers

+ Creation of a Thematic Map (Choropleth) by combining vector data

---
class:center,middle,inverse
# Basic Map Elements

---
class:center,middle

## Points  
![:width 50%](data:image/png;base64,#images/point_feature.png)
.caption[http://docs.qgis.org/2.8/en/docs/gentle_gis_introduction/vector_data.html#overview]

---
class:center,middle

## Lines
![:width 50%](data:image/png;base64,#images/polyline_feature.png)
.caption[http://docs.qgis.org/2.8/en/docs/gentle_gis_introduction/vector_data.html#overview]

---
class:center,middle

## Polygons
![:width 50%](data:image/png;base64,#images/polygon_feature.png)
.caption[http://docs.qgis.org/2.8/en/docs/gentle_gis_introduction/vector_data.html#overview]

---
## Shapefiles
--
        
+ Basic file for storing map elements
--

+ Stores spatial data, like points, lines, and polygons
--

+ Multiple files comprise a "shapefile"

![img-center-80](data:image/png;base64,#images/file_struct1.png)
---
# Reworking of Shapefiles

+ Spatial (vector) data often can be downloaded, or created using digitalisation of scanned maps (last session)

+ Sometimes it is necessary to combine informations (geometries) for the desired result

+ for theses tasks there are geometry processing tools

![](data:image/png;base64,#images/geo_processing01.png)
---

## Let's Get Started
1. [Click this link](https://github.com/BernCoDALab/gia/raw/main/lectures/05/data/kantonsgrenzen.zip) and download the cantonal borders of Switzerland (shapefile) to your desktop
2. [Click this link](https://github.com/BernCoDALab/gia/raw/main/lectures/05/data/seen.zip) and download the lakes of Switzerland (shapefile) to your desktop
3. [Click this link](https://github.com/BernCoDALab/gia/raw/main/lectures/05/data/14c_daten.csv) and download some 14C data (csv) also to your desktop
4. Open QGIS
5. Start a new project and Add all layers

---
## Adding a CSV Layer...Reminder

+ csv data are 'delimited text' data, so use the approriate import tool
+ the coordinates are store as lat/lng in columns named accordingly
  + lat -> *Latitude:* a geographic coordinate that specifies the **north–south** position of a point on the Earth's surface -> y-coordinate
  + lng -> *Longitude:* a geographic coordinate that specifies the **east–west** position of a point on the Earth's surface -> x-coordinate
+ latitude/longitude -> World Geodetic System: WGS 84 -> **EPSG 4326**

.center[![:width 50%](data:image/png;base64,#images/csv_import01.png)]

---
## Tasks for today

+ Extract the 14C data for Switzerland
+ Count how many data are in lakes and within 1 km around lakes
+ visualise the mean age of 14C dates per canton

.center[![:width 50%](data:image/png;base64,#images/tasks01.png)]

+ first, we need a polygon of Switzerland (or create one ourself)

---
## Vector overlay
.pull-left[
Combine different layers for to produce a combined result

### Multiple Options
+ Clip (Intersection)
+ Difference (Substraction)
+ Symmetric Difference (Disjunction)
+ Union

![](data:image/png;base64,#images/jccb_0503.png)
.caption[https://moderndilettante.wordpress.com/category/boole/]
]
.pull-right[![](data:image/png;base64,#images/geo_processing01.png)]

---
## Combining lakes with cantons

.pull-left[
The cantonal borders exclude the lakes! Check yourself...

We need to combine the lakes with the borders shapefile!

.tiny[*Not strictly necessary for our task at hand, but serves didactic purpose...*]

**Which algorithm to use?**

![](data:image/png;base64,#images/jccb_0503.png)
.caption[https://moderndilettante.wordpress.com/category/boole/]
]
.pull-right[
![](data:image/png;base64,#images/geo_processing02.png)

![](data:image/png;base64,#images/geo_processing03.png)
]

---
## Union

+ Select Union from Vector > Geometry-Processing
+ Select the cantonal borders as one layer
+ Select the lakes as second layer
+ Temporary Layer is fine, so no need to specify storage location
+ Check 'Open after success' checkbox
+ Click 'Run'

.center[
![:width 75%](data:image/png;base64,#images/geo_processing04.png)
]
---
## Union Result

+ A layer containing all polygons from both layers
+ You can check in the attribute table:

.center[
![:width 75%](data:image/png;base64,#images/geo_processing05.png)
]

---
## Dissolve

.pull-left[
+ We need only one polygon -> Combining all polygons to one
+ use disolve tool (Vector > Geometry-Processing)
+ Select Union Layer as Input Layer
+ Temporary Layer as output is fine
+ Click 'Run'
]

.pull-right[
![:width 75%](data:image/png;base64,#images/geo_processing06.png)

![:width 75%](data:image/png;base64,#images/geo_processing07.png)
]

---
## Wrong or unnecessary Vertices

If we zoom in, some Structure within our polygon resulted from non-matching spatial data:

.right[
![:width 75%](data:image/png;base64,#images/geo_processing08.png)
]

---
## Edit/Remove wrong or unnecessary Vertices

.pull-left[
+ Click on 'Toggle Edit Mode' ![](data:image/png;base64,#images/geo_processing09a.png)
+ Click on Vertices Tool ![](data:image/png;base64,#images/geo_processing09b.png)
+ Select unnecessary vertices and remove
+ Save by and 'Toggle Edit Mode' ![](data:image/png;base64,#images/geo_processing09a.png)

*in the same tool, you also can move points of your polygons to different locations...*
]

.pull-right[
![](data:image/png;base64,#images/geo_processing09.png)
]
---
## Toolbox

.pull-left[
For our next task, we need a tool that is not available from the menu

For this, we activate the toolbox sidebar/window ![](data:image/png;base64,#images/geo_processing10a.png)

This toolbox contains all menu tools + a lot more

It also features a search bar for fast and easy access
]

.pull-right[
![](data:image/png;base64,#images/geo_processing10.png)
]

---
## Extract points based on Polygon (1)

To extract the 14C data contained within the 'Switzerland'-polygon, we access the 'Extract by Position' tool. Start typing in the search field to find it!

![](data:image/png;base64,#images/geo_processing11.png)

---
## Extract points based on Polygon (2)

.pull-left[
+ Select the 14c-layer as layer to extract from
+ Select 'are contained' as Option
+ Select the 'Dissolved' Layer as source
+ Temporary Layer might be fine
+ Click on 'Run'
+ The resulting Layer should only contain the 14C Data within Switzerland

+ Finally save your layer as shapefile!
]

.pull-right[
![](data:image/png;base64,#images/geo_processing11.png)
![](data:image/png;base64,#images/geo_processing12.png)
]
---
## Check the number of 14C Dates within lakes (1)

*(Spatial Query)*

Lets start with our first analytical procedure: How many 14C dates are comming from lakes?

To check this, we need the lakes and the 14C Data:

![:width 60%](data:image/png;base64,#images/count_in_polygon01.png)
---
## Check the number of 14C Dates within lakes (2)

This time, the tool is accessible via the menue 'Vector > Analytical tools':

![:width 48%](data:image/png;base64,#images/count_in_polygon02.png) ![:width 48%](data:image/png;base64,#images/count_in_polygon03.png)

+ select the lakes layer as polygon layer
+ select the 14C layer as points layer
+ (you can specify a weighting and a classification field, too)
+ the name of the resulting count column is 'NUMPOINTS', that might be ok for now
+ temporary layer might be ok for now
+ click on 'Run'

---
## Check the number of 14C Dates within lakes (2)

A new layer is created that contains the original information from the lakes layer + the number of 14C data as 'NUMPOINTS'. You can check in the attribute table.

![](data:image/png;base64,#images/count_in_polygon04.png)

---
## Check the number of 14C Dates within 1km around lakes (1)

.pull-left[
To check, how many 14C Data are within a buffer of 1 km around the lakes, we have to create a buffer first.

+ Select Buffer from 'Vector > Geometry processing tools'
+ Select the lakes layer as input layer
+ Select 1km as distance
+ most of the other options refer to the shape of the buffer... we can leave them to default now
+ the checkbox 'dissolve' merges all connected buffers, we do not want that here
+ Click on 'Run'
]

.pull-right[
![:width 90%](data:image/png;base64,#images/buffer01.png)

![:width 90%](data:image/png;base64,#images/buffer02.png)
]

---
## Check the number of 14C Dates within 1km around lakes (2)

+ The resulting Vector layer contains all informations of the lakes layer, only that the lakes are 1km enlarged
+ You can now count points in polygons like before, but you should select the 'buffered' layer as your input
+ The number of 14C dates are now much larger per lake

---
## Calculate mean 14C age per canton (1)

Using GIS, you can excecute statistical procedures and visualise the result in a Choropleth.

(the German term for this is also Choropleth...)

![:width 50%](data:image/png;base64,#images/median_reineinkommen.gif)

#####Source: http://www.estv2.admin.ch/

---
## Calculate mean 14C age per canton (2)

*(Spatial Join)*

To start, we need the cantonal borders and the 14C data and the Toolbox ![](data:image/png;base64,#images/geo_processing10a.png).

The command is 'Join attributes by location (summary)'

![:width 80%](data:image/png;base64,#images/spatial_join01.png)
---
## Calculate mean 14C age per canton (3)

+ Select the cantonal borders as input layer
+ Select the 14C data as joined layer
+ Select intersection as condition

![:width 80%](data:image/png;base64,#images/join_summary01.png)
---
## Calculate mean 14C age per canton (4)

+ in Summary fields, select bp (uncalibrated 14C date) as field
+ in Calculated Summaries, select 'mean' as statistic
+ temporary layer might be ok
+ press 'Run'

![:width 48%](data:image/png;base64,#images/join_summary02.png) ![:width 48%](data:image/png;base64,#images/join_summary03.png)

---
## Calculate mean 14C age per canton (5)

In the Attribute Table, now you can see the mean bp age per canton

![](data:image/png;base64,#images/join_summary04.png)

---
## Visualise mean 14C age per canton

Now we can color the cantons according to the mean BP age

+ Open the **Properties** of the layer (right click on the layer) and Select there the **Symbology** tab

+ Select 'Graduated' as symbol and Classify the values

+ click on 'OK'

+ Now you should be able to see the age by color

![:width 48%](data:image/png;base64,#images/join_summary05.png) ![:width 48%](data:image/png;base64,#images/join_summary06.png)

---
## What We've Covered
--

+ Editing and joining Geometries
--

+ Extract features based on location in relation to other vector data
--

+ Extract and summarise information based on location in relation to other vector data

---

## Homework

+ Get the [north](https://github.com/BernCoDALab/gia/raw/main/lectures/05/data/kreise_bern_nord.zip) and the [south part](https://github.com/BernCoDALab/gia/raw/main/lectures/05/data/kreise_bern_sued.zip) of the Kanton of Bern
+ Unite them to get a map of the whole Kanton
+ Get the [Location of archaeological sites of the Kanton of Bern](https://github.com/BernCoDALab/gia/raw/main/lectures/05/data/fundstellen_bern.zip)
+ Count the number of sites in each 'Amtsbezirk'
+ Make a Map showing the number of sites color coded
+ Send me the map

---
class: inverse, middle, center
# Any questions?

![:width 20%](data:image/png;base64,#images/pointless_mountain.png)

.caption[Source: https://www.instagram.com/sadtopographies]

.footnote[
.right[
.tiny[
You might find the course material (including the presentations) at

https://github.com/BernCoDALab/gia

You can see the rendered presentations at

https://berncodalab.github.io/gia

You can contact me at

<a href="mailto:martin.hinz@unibe.ch">martin.hinz@unibe.ch</a>
]
]
]