plotnine_contest.qmd

---
title: "Plotnine Contest"
format:
  html:
    toc: true
    embed-resources: true
    other-links:
      - text: The Himalayan Database
        href: https://www.himalayandatabase.com/
      - text: Tidy Tuesday Project
        href: https://github.com/rfordatascience/tidytuesday/blob/master/data/2020/2020-09-22/readme.md  
---

# Eight-thousanders in Nepal

For my submission, I use data from [The Himalayan Database](https://www.himalayandatabase.com/). I found the csv in the [Tidy Tuesday Project](https://github.com/rfordatascience/tidytuesday/blob/master/data/2020/2020-09-22/readme.md). 

# Data wrangling

```{python}

from plotnine import *
import geopandas
import geocoder
import pandas as pd
from siuba import *
```

## Filter the relevant mountains

There are [14 eight-thousanders](https://en.wikipedia.org/wiki/Eight-thousander) in the world. Eight of them are (partially) in Nepal and their ascents are well-documented in The Himalayan Database. Those mountains will be the focus of my analysis and data visualization. 

```{python}

# Import the csv from the Tidy Tuesday project
members = pd.read_csv(
    "https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-09-22/members.csv"
)

peaks = pd.read_csv(
    "https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-09-22/peaks.csv"
)
```

```{python}

# Filter for the eight-thousanders
df01 = peaks >> filter(_.height_metres > 8000)

# I want the ascents on the individual level (not on expedition-level)
df02 = inner_join(df01, members, by=["peak_id", "peak_name"])
```

```{python}

# I only want the main peaks of the eight-thousanders (not the subsidiary peaks)
main_peaks = (
    df02
    >> count(_.peak_name)
    >> filter(_.n > 1000)
)
```

```{python}

# I want to focus on the more recent years when there were coming more and more big commercial expeditions
df03 = (
    df02 
    >> filter(_.peak_name.isin(main_peaks["peak_name"]))
    >> filter(_.year >= 2009)
)
```

```{python}

# I am interested in the success differentiated by whether oxygen was used or not

oxygen_use_sucess_01 = (
    df03
    >> count(_.peak_name, _.oxygen_used, _.success)
    >> mutate(peak_name = _.peak_name.str.replace("I", ""))
)
```

## Geocoding

```{python}

# I need the latitude and longitude of the mountains
lat = []
lng = []

for peak in oxygen_use_sucess_01["peak_name"]:
    latlng = geocoder.arcgis(peak).latlng
    lat.append(latlng[0])
    lng.append(latlng[1])

oxygen_use_sucess_01["lat"] = lat
oxygen_use_sucess_01["lng"] = lng
```

```{python}

# Move some mountains a little bit on the map so that they do not overlap that much
oxygen_use_sucess_02 = (
    oxygen_use_sucess_01 
    >> mutate(lng=case_when({
        _.peak_name == "Makalu": _.lng + 0.15,
        _.peak_name == "Lhotse": _.lng + 0.15,
        _.peak_name == "Cho Oyu": _.lng - 0.15,
        True: _.lng,
    }))
    >> rename(Oxygen = _.oxygen_used, Success = _.success)
)
```

## Geopandas

```{python}

# Load the world shapefile data
world = geopandas.read_file(
    "https://github.com/geopandas/geopandas/raw/v0.9.0/geopandas/datasets/naturalearth_lowres/naturalearth_lowres.shp"
)

# Filter for Nepal
nepal = world >> filter(_.name == "Nepal")
```

# Creating the plot

```{python}

p = (
    ggplot(oxygen_use_sucess_02)
    + geom_map(nepal, fill="#d0d0d0", size=0)
    + geom_point(aes(x="lng", y="lat", size="n", color="peak_name"))
    + facet_wrap(["Oxygen", "Success"], labeller="label_both")
    + coord_fixed()
    + theme_void()
    + labs(
        size="Number of climbers",
        color="Mountain",
        title="Success on > 8,000 m peaks in Nepal without bottled oxygen is rare",
        subtitle="Everest has only been summited 57 times without bottled oxygen (and 5,903 times using bottled oxygen)",
        caption="Data: The Himalayan Database\nYears: 2009 - 2019",
    )
    + theme(
        plot_background=element_rect(fill="#EBF2FF"),
        legend_position="bottom",
        plot_title=element_text(ha="center", size=19),
        plot_subtitle=element_text(ha="center"),
    )
    + scale_size_continuous(
        range=[1, 20],
        breaks=[50, 100, 1000, 3000, 5000],
        labels=["50", "100", "1,000", "3,000", "5,000"],
    )
)
```

```{python}

#p.save("mathis_plotnine_contest.png", height = 8, width = 10, dpi=600)
```

# My submission

The plot can be found in @fig-submission. **For the best viewing experience, open it in an external tab.**

![Plotnine Contest Submission Mathis](mathis_plotnine_contest.png){#fig-submission}