M1 revision spring 2024 (#53)

* Update mkdocs.yml

Changed repo_url to url of the fork

* add toolbox intro documents

* add draft for geopython quickstart tutorial

* add toolbox intro to mkdocs.yml

* update mkdocs.yml

* add tutorials for geopandas, rasterio and (basic) xarray

* add tutorials for geopandas, rasterio and xarray to mkdocs.yml

* revise conda.md

* revise jupyter.ipynb

* add dissolve example, add custom maps example

* extend rasterio tutorial by examples for plotting and multi-band handling

* conclude xarray tutorial

* format link

* add figure illustrating (non-)stationarity

* correct figure alt description

* geopandas dissolve example: provide dissolve column via dictionary to aggfunc

* update STAC endpoint and collection

* update and clean theme 2 main document

* add subheading

* update input data path

* remove single-image part of the tutorial (now part of the toolbox intro)

* Add toolbox figures (snakes, rasterio, xarray)

* Fix small issues (broken links and redundant code)

* Replace figure (snakes)

* Delete caption

* Rename figure

* Remove toolbox intro to be able to merge only bug fixes and typo fixes now

* Update mkdocs.yml (re-adjust repo_url to main course url)

re-adjust repo_url to main course url instead of the forked course url

course/module1/01_principles_of_remote_sensing_time_series/01_principles_of_remote_sensing_time_series.md

@@ -21,7 +21,7 @@ At the end of this theme, you will know concepts and typical properties of time
 
 ### What is a time series?
 
-A time series is *a set of observations that are ordered by time* (i.e., by their temporal occurrence; [Castle and Hendry 2019](https://doi.org/10.1007/978-3-030-21432-6)). That means the observations have a clear and unchangeable chronology and distance to each other. The chronology of observations may been seen like neighbourhood and topology in spatial data analysis.
+A time series is *a set of observations that are ordered by time* (i.e., by their temporal occurrence; [Castle and Hendry 2019](https://doi.org/10.1007/978-3-030-21432-6)). That means the observations have a clear and unchangeable chronology and distance to each other. The chronology of observations may be seen like neighbourhood and topology in spatial data analysis.
 
 Time series are recorded at different frequencies (number of measurements in given time period), which are often expressed through a measurement interval (i.e. the temporal spacing between consecutive measurements). This can be a (more or less) fixed interval (e.g. hourly, daily, every five days, …) or an irregular one. The moment in time when a measurement is recorded is usually referenced as a time stamp (e.g. *June 5th, 2019 at 7:00 am*).
 
@@ -55,6 +55,9 @@ The persistent patterns in the temperature time series can be clearly seen if we
 
 A time series where the mean and variance of the observations stay constant over time is considered *stationary* (see [Castle and Hendry (2019](https://doi.org/10.1007/978-3-030-21432-6), p. 10) for more detailed explanations and examples). In many environmantal and socio-economic time series this is not the case (*non-stationarity*). Staying with the air temperature example, it matters a lot if temperature on a specific day of the year (DOY) was measured in the 1920ies or very recently, with climate change having now strongly affected the temperatures to be expected with a certain probability. Non-stationarity can be related to gradual developments (like atmospheric CO<sub>2</sub> increase) but also to sudden events (like volcanic eruptions), which cause shifts in the means and variances of the data. If undetected, non-stationarity is a problem for empirical forecasting and for an analysis of relations between variables by their temporal development ([Castle and Hendry 2019](https://doi.org/10.1007/978-3-030-21432-6)). Therefore, it can be advisable to check the time series for stationarity, e.g. using the Kwiatkowski-Phillips-Schmidt-Shin test ([Kwiatkowski et al. 1992](https://doi.org/10.1016/0304-4076(92)90104-Y)) or by visual inspection of plotted values.
 
+<img src="media/non-stationarity.png" title="Examples of a stationary time series with constant mean and variance and (b) three non-stationary time series" width="800">
+
+*Examples of (a) a stationary time series with constant mean and variance and (b) three non-stationary time series in the form of a shift in mean (trend and step change) and a shift in variance. Solid and dashed black lines represent the mean and the variance of the time series, respectively. (figure by [Slater et al. 2021](https://doi.org/10.5194/hess-25-3897-2021)/ [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)).*
 
 #### Trend detection and removal