import folium
import numpy as np
import warnings
from folium.plugins import FloatImage
warnings.filterwarnings('ignore')
warnings.filterwarnings(action='once')

import pandas as pd
import geopandas as gpd
import rasterio as rs
import matplotlib.pyplot as plt
from matplotlib import cm
import matplotlib

import plotly.plotly as py
import plotly.graph_objs as go


%matplotlib inline

Markdown Monster icon

Reunion a hot spot of erosion on Earth

Storm-induced precipitation variability control of long-term erosion.

E. Gayer, L. Michon, P. Louvat, J. Gaillardet. EPSL 2019-DOI: 10.1016/j.epsl.2019.04.003

Erosion is often treated as a continuous process, yet it occurs through discrete events such as floods and landslides of variable magnitude and periodicity. It has also long been expected to be strongly dependent on precipitation, however, the influence of temporal rainfall variability upon long-term evolution of landscapes remains unclear. Here we report high erosion rates (0.8 to ~10 mm yr^-1 over ~70 ka) estimated from paleovolcanic reconstructions across a steep rainfall gradient on Reunion Island, which show that long-term erosion rates are influenced by the cyclone-induced variability of precipitation. Geostatistical analysis of 30 years of daily rainfall records reveals that erosion rates are high where the local climate is the driest and where the difference in intensity between extreme rainfall events and prevailing precipitation is the strongest. This implies that the intrinsic variability of precipitation impacts landscape evolution not only through extreme meteorological events, but also through background rainfall-induced parameters such as humidity and dryness, which modulate the erosion threshold of the Earth's Critical Zone.

-- The map below allows to choose between --

The erosion rate distribution across Reunion
The current topgraphy
The modeled topography of the island ~70kyrs ago reconstructed from paleosurfaces
The mean annual precipitation distribution (1981-2010)
The mean annual clyclonic precipitation distribution (1981-2010)
the coefficient of variation of the daily precipitation (standard deviation / average)
The trajectories of the 13 most important Cyclone betweens 1980 and 2010 (with their wind speed)

[** Hover over the area with the mouse cursor for additional information]

m = folium.Map([-21.15, 55.5], tiles='CartoDB positron', zoom_start=10)

#################
#-CARTE EROSION--
#################

#### --- Donnees erosions 7 bassins ----------
df_ero=pd.read_csv('/Users/egayer/Desktop/test_siteJupCarte/TableauErosionBassin_ordreSHP.csv')
gdf_7bassins = gpd.read_file('/Users/egayer/Desktop/test_siteJupCarte/WGS84_LatLon/7StudiedBassin_WGS84LatLon.shp')

#### --- Echelle de Couleur des tax erosion/ Valeur hex ----------
cmap = matplotlib.cm.get_cmap('autumn_r')
minero=df_ero['ErosionK'].min()
maxero=df_ero['ErosionK'].max()

value01=(df_ero['ErosionK']-minero)/(maxero-minero)
listcol=[]
for i in value01:
    rgb = cmap(i)[:3] # will return rgba, we take only first 3 so we get rgb
    listcol+=[matplotlib.colors.rgb2hex(rgb)]

################## -- CARTE -- ##################
fg=folium.FeatureGroup(name='Watershed Erosion', show=True)
m.add_child(fg)
for i in range(len(gdf_7bassins)):
    x,y=gdf_7bassins.iloc[i].geometry.exterior.coords.xy
    bassname=gdf_7bassins['Name'].iloc[i]
    erosion=df_ero['ErosionK'][df_ero['Bass']==bassname]
    folium.vector_layers.Polygon(locations=zip(y,x),
                            #popup=bassname +'\n'+str(erosion.values[0])+ ' t/km2/yr',
                            #tooltip=bassname,
                            tooltip=bassname +': '+str(erosion.values[0])+ ' t/km2/yr',
                            fillOpacity=0.7,
                            color='white',
                            weight=1,
                            #fill_color=listcol[i]).add_to(m)
                            fill_color=listcol[i]).add_to(fg)

lim_forall=[[-21.417461113593202, 55.19145732801233], [-20.844276883025962, 55.86336961788544]]    

#################
#-CARTE TOPO ACTUELLE --
#################
Topo_now = plt.imread('/Users/egayer/Desktop/test_siteJupCarte/Run25mWGS84LatLon.png')
folium.raster_layers.ImageOverlay(
    image=Topo_now,
    bounds=lim_forall,
    opacity=0.5,
    name='Current Topo Hillshade',
    show=False
).add_to(m)

#################
#-CARTE TOPO Init --
#################
Topo_init = plt.imread('/Users/egayer/Desktop/test_siteJupCarte/Zrun25mWGS84LatLon_Topo_InitCD.png')
folium.raster_layers.ImageOverlay(
    image=Topo_init,
    bounds=lim_forall,
    opacity=0.5,
    name='Initial (~70ka ago) Topo Hillshade',
    show=False
).add_to(m)

#################
#-CARTE Pluvio Moy --
#################
Pluvio_Moy = plt.imread('/Users/egayer/Desktop/test_siteJupCarte/PluvioAnnuelle_Totale_25WGS84LatLon_LM_ssRelief.png')
folium.raster_layers.ImageOverlay(
    image=Pluvio_Moy,
    bounds=lim_forall,
    opacity=0.5,
    name='Mean Annual Precipitation',
    show=False
).add_to(m)

#################
#-CARTE Pluvio Cyclo --
#################
Pluvio_Cyclo = plt.imread('/Users/egayer/Desktop/test_siteJupCarte/PluvioCyclone_Totale1981_2010_25WGS84LatLon_LM.png')
folium.raster_layers.ImageOverlay(
    image=Pluvio_Cyclo,
    bounds=lim_forall,
    opacity=0.5,
    name='Mean Annual Cyclonic Precipitation',
    show=False
).add_to(m)


#################
#-CARTE Cvp --
#################
Coeff_Var = plt.imread('/Users/egayer/Desktop/test_siteJupCarte/zCoeffVar_Journa_gridNN_WGS84LatLon.png')
folium.raster_layers.ImageOverlay(
    image=Coeff_Var,
    bounds=lim_forall,
    opacity=0.5,
    name='Coefficient of variation',
    show=False
).add_to(m)

#################
#-Trajectoire des Cyclo --
#################
fileD=['/Users/egayer/MATLAB-Stuff/Reunion/PapierErosion/PluvioData/Trajectoire11Cyclones/Florine81Traj.txt',
'/Users/egayer/MATLAB-Stuff/Reunion/PapierErosion/PluvioData/Trajectoire11Cyclones/Erinesta86Traj.txt',
'/Users/egayer/MATLAB-Stuff/Reunion/PapierErosion/PluvioData/Trajectoire11Cyclones/Clotilda87Traj.txt',
'/Users/egayer/MATLAB-Stuff/Reunion/PapierErosion/PluvioData/Trajectoire11Cyclones/Firinga89Traj.txt',
'/Users/egayer/MATLAB-Stuff/Reunion/PapierErosion/PluvioData/Trajectoire11Cyclones/Colina93Traj.txt',
'/Users/egayer/MATLAB-Stuff/Reunion/PapierErosion/PluvioData/Trajectoire11Cyclones/Hollanda94Traj.txt',
'/Users/egayer/MATLAB-Stuff/Reunion/PapierErosion/PluvioData/Trajectoire11Cyclones/Connie2000Traj.txt',
'/Users/egayer/MATLAB-Stuff/Reunion/PapierErosion/PluvioData/Trajectoire11Cyclones/Ando2001Traj.txt',
'/Users/egayer/MATLAB-Stuff/Reunion/PapierErosion/PluvioData/Trajectoire11Cyclones/Dina2002Traj.txt',
'/Users/egayer/MATLAB-Stuff/Reunion/PapierErosion/PluvioData/Trajectoire11Cyclones/Diwa2006Traj.txt',
'/Users/egayer/MATLAB-Stuff/Reunion/PapierErosion/PluvioData/Trajectoire11Cyclones/Gamede2007Traj.txt']

Lat=[]
Lon=[]
Vit=[]
maxVit=[]
minVit=[]

fg2=folium.FeatureGroup(name='13 Biggest Cyclones 1981-2010', show=False)
m.add_child(fg2)
for idx in range(len(fileD)):
    #print idx
    #print fileD[idx]
    LatI,LonI,VitI=np.loadtxt(fileD[idx],unpack=True)
    LatI=-LatI/10.
    LonI=LonI/10.
    #maxVitI=np.max(VitI)
    #minVitI=np.min(VitI)
    #maxVit=maxVit+[maxVitI]
    #minVit=minVit+[minVitI]    
    #Lat=Lat+[LatI]
    #Lon=Lon+[LonI]
    #Vit=Vit+[VitI]
    color_line = folium.features.ColorLine(
    positions=zip(LatI, LonI),
    colors=VitI,
    #colormap=['y', 'orange', 'r'],
    colormap=['cyan', 'magenta'],
    weight=3)

    color_line.add_to(fg2)

folium.LayerControl().add_to(m)

#################
#-Legende -- (PNG transforme en base64 par https://www.base64-image.de)
#################
#leg= ('data:image/png;base64,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')
leg= ('data:image/png;base64,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')

FloatImage(leg, bottom=2, left=2).add_to(m)

m

#<img src="ColorBar_map.png" alt="legends">

Here we used paleovolcanic reconstructions and statistical analysis of 30 years of rainfall records to show that high long-term erosion rates on Reunion Island (0.8 to ~10 mm yr-1 over ~70 ka)are correlated with the cyclone-driven variability of precipitation, rather than with mean annual precipitation (Fig. 1 and 2 below). We show that estimated erosion rates are high where the difference in intensity between extreme rainfall and prevailing precipitation is the strongest indicating that storm control on erosion is influenced by factors that depend on background rainfall regime.

[** Hover over the glyphs with the mouse cursor for name of the watershed]

import pandas as pd
from bokeh.plotting import figure, show
from bokeh.models import ColumnDataSource, Whisker, TeeHead,ColorBar,Legend
from bokeh.io import output_notebook
from bokeh.palettes import RdYlBu, Spectral6
from bokeh.transform import linear_cmap
from bokeh.layouts import column, row

# Faire les graphs eroion vs pluvio / cyclo/ CVp

df_values=pd.read_csv('/Users/egayer/Desktop/test_siteJupCarte/TableauErosionPluvioCycloCVP.csv', index_col=0)

#df_values.head()

df_values_forbokeh=df_values[['Emass','errmass','MeanAnnuel_Pluvio_surBassin_mm_a','MeanCyclo_Pluvio_surBassin_mm_a']]
df_values_forbokeh=df_values_forbokeh/1000.

df_values_forbokeh.columns=['Emass_Kt','errmass_Kt','MeanAnnuel_Pluvio_surBassin_m_a','MeanCyclo_Pluvio_surBassin_m_a']

df_values_forbokeh['Emass_Kt_upper']=df_values_forbokeh['Emass_Kt']+df_values_forbokeh['errmass_Kt']
df_values_forbokeh['Emass_Kt_lower']=df_values_forbokeh['Emass_Kt']-df_values_forbokeh['errmass_Kt']


#df_values_forbokeh.head()

df_values_forbokeh['CVJourna']=df_values['CVJourna']
df_values_forbokeh['err_CVJourna']=df_values['err_CVJourna']

#df_values_forbokeh.head()

# PLOT P

sourcep=ColumnDataSource(df_values_forbokeh)
sourcep.add(df_values_forbokeh.index, 'index')
HoverName = [("Watershed","@index"),]
p = figure(plot_width= 500, plot_height=(500*2/3), title='Fig.1 Basin average Mean Precipitation',x_range=(0, 7), y_range=(0, 35),tooltips=HoverName)
color_mean="deepskyblue"
color_cyclo="deeppink"

pm=p.scatter(x='MeanAnnuel_Pluvio_surBassin_m_a', y='Emass_Kt', size=18, source=sourcep,color=color_mean,marker="circle",
            line_color="dodgerblue", line_width=2)
pc=p.scatter(x='MeanCyclo_Pluvio_surBassin_m_a', y='Emass_Kt', size=20, source=sourcep,color=color_cyclo,marker="triangle",
            line_color="mediumvioletred", line_width=2)
p.add_layout(Whisker(source=sourcep, base="MeanAnnuel_Pluvio_surBassin_m_a", upper="Emass_Kt_upper", lower="Emass_Kt_lower",
                     line_color=color_mean,upper_head=TeeHead(line_color=None), lower_head=TeeHead(line_color=None)))
p.add_layout(Whisker(source=sourcep, base="MeanCyclo_Pluvio_surBassin_m_a", upper="Emass_Kt_upper", lower="Emass_Kt_lower",
                    line_color=color_cyclo,upper_head=TeeHead(line_color=None), lower_head=TeeHead(line_color=None)))

legend = Legend(items=[
    ("Mean"   , [pm]),
    ("Cyclonic" , [pc])], location="center")

p.add_layout(legend, 'right')
p.legend.border_line_color = None

#Titre axes
p.title.text_font_size = '12pt'
p.xaxis.axis_label = "Basin average Mean Precipitation (m/yr)"
p.yaxis.axis_label = "Erosion Rates (Kt/km2/yr)"
p.axis.axis_label_text_font_style = 'normal'

# GRID STYLE
p.ygrid.grid_line_color = "silver"
p.ygrid.grid_line_alpha = 1
p.ygrid.grid_line_dash = [2, 3]
p.xgrid.grid_line_color = "silver"
p.xgrid.grid_line_alpha = 1
p.xgrid.grid_line_dash = [2, 3]

# AXE STYLE
p.xaxis.axis_line_color = None
p.yaxis.axis_line_color = None

p.outline_line_color = "silver"
p.outline_line_dash=[2,3]

p.xaxis.major_tick_line_color = "silver"
p.xaxis.minor_tick_line_color = None
p.yaxis.major_tick_line_color = "silver"
p.yaxis.minor_tick_line_color = None


p.axis.major_tick_out = 10
p.axis.minor_tick_out = 5

# REMOVE TOOLBAR
p.toolbar.logo = None
p.toolbar_location = None

#################
# PLOT O
sourceo=ColumnDataSource(df_values_forbokeh)
sourceo.add(df_values_forbokeh.index, 'index')

o = figure(plot_width= 450, plot_height=(500*2/3), title='Fig.2 Basin average Coefficient of variation',x_range=(0, 7), y_range=(0, 35),tooltips=HoverName)
source=ColumnDataSource(df_values_forbokeh)
mapper = linear_cmap(field_name="CVJourna", palette=RdYlBu[8] ,low=df_values_forbokeh['CVJourna'].min() ,high=df_values_forbokeh['CVJourna'].max())

o.scatter(x='CVJourna', y='Emass_Kt', size=18, source=sourceo,color=mapper,marker="circle",
            line_color="black", line_width=2)

o.add_layout(Whisker(source=sourceo, base="CVJourna", upper="Emass_Kt_upper", lower="Emass_Kt_lower",
                     line_color="black",upper_head=TeeHead(line_color=None), lower_head=TeeHead(line_color=None)))


color_bar = ColorBar(color_mapper=mapper['transform'], width=8,  location=(0,0), title="CV-p")

o.add_layout(color_bar, 'right')

#Titre axes
o.title.text_font_size = '12pt'
o.xaxis.axis_label = "CV of Precipitation"
o.yaxis.axis_label = "Erosion Rates (Kt/km2/yr)"
o.axis.axis_label_text_font_style = 'normal'


# GRID STYLE
o.ygrid.grid_line_color = "silver"
o.ygrid.grid_line_alpha = 1
o.ygrid.grid_line_dash = [2, 3]
o.xgrid.grid_line_color = "silver"
o.xgrid.grid_line_alpha = 1
o.xgrid.grid_line_dash = [2, 3]

# AXE STYLE
o.xaxis.axis_line_color = None
o.yaxis.axis_line_color = None

o.outline_line_color = "silver"
o.outline_line_dash=[2,3]

o.xaxis.major_tick_line_color = "silver"
o.xaxis.minor_tick_line_color = None
o.yaxis.major_tick_line_color = "silver"
o.yaxis.minor_tick_line_color = None


o.axis.major_tick_out = 10
o.axis.minor_tick_out = 5

# REMOVE TOOLBAR
o.toolbar.logo = None
o.toolbar_location = None

output_notebook(hide_banner=True)
show(row(p,o))

Fig. 1 shows the relationship between basin-averaged erosion rates and 1) upstream-averaged mean annual precipitation rates (circles); 2) upstream-averaged mean annual cyclonic precipitation rates (triangles), displaying no major gradient across the 7 watersheds.

Fig. 2 shows the basin-averaged erosion rates as a function of CV-p

Estimation of the consequences of changes in climatic variability on landscape evolution is contingent upon specific processes and poorly constrained boundary conditions. However, our results demonstrate that climatic variability increases erosion rates and thus support the general assumption that climate shifts toward arid conditions (dominated by extreme events) may increase erosion (Molnar et al., 2006; Tucker and Bras, 2000). Furthermore, the future shifts of tropical-cyclone activity highlighted by some climatic models (e. g. Redmond et al., 2015) may expose new coastal areas (Darby et al., 2016) to high rainfall variability and erosion rates. Climatic variability is a factor that should be considered when attempting to relate climate and long-term erosion rates, not only to interpret stratigraphic records but also to predict the potential impacts of future climate changes.