🎯要點(diǎn)

1. 算法區(qū)分不同水循環(huán)數(shù)據(jù)類型：地下水、河水、降水、氣溫和其他，并使用相應(yīng)標(biāo)準(zhǔn)化降水指數(shù)、標(biāo)準(zhǔn)化地下水指數(shù)、標(biāo)準(zhǔn)化河流水位指數(shù)和標(biāo)準(zhǔn)化降水蒸散指數(shù)。
2. 繪制和計(jì)算特定的時(shí)間序列比較統(tǒng)計(jì)學(xué)相關(guān)性。
3. 使用相關(guān)矩陣可視化集水區(qū)和顯示空間信息。

🍪語言內(nèi)容分比

🍇Python地圖表示

Altair 是一個(gè)基于 Vega 和 Vega-Lite（來自 JS）的聲明式圖形可視化庫，而 Vega 和 Vega-Lite 又基于 D3.js。它允許將交叉地圖與其他類型的圖形相關(guān)聯(lián)，如以下示例所示：

import altair as alt
url_geojson = 'mallorca_geoson.json'
data_geojson = alt.Data(url=url_geojson, format=alt.DataFormat(property=' features',type='jjson'))
mallorca = alt.Chart(data_geojson).mark_geoshape(stroke="gray", color="white", strokeWidth=0.5)
#D Data
coord_mask = (pluvio.COORD_X >= 2.2) & (pluvio.COORD_X <= 3.5)&\(pluvio.COORD \overline{Y}>=38.\theta)&(pluvio.COORD_
mallorca_pluvio = pluvio.\??oc[coord_mask, : ]
#Plots
brush = alt.selection_interval(encodings=["longitude", "latitude"], empty=False)
pluviometros = alt.Ch?rt(mallorca_pluvio) \.mark circle(size=50)\.encode(longitude =' COORD X:Q',latitude='COORD \Y:Q',tooltip=['NOMBRE:N', 'ALTITUD:Q'],color=alt.condition(brush, alt.value("red"), alt.value("lightgray"))) \.project("equalEarth") \.properties(width=500, height=350) \. add params(brush)
bars = alt.Chart(mallorca_pluvio) \.mark_bar() \.encode(x=alt.X("ALTITUD:Q"),bin(extent=[0,700]),y=" count (ALTITUD):0",color=alt.value("steelblue"))left_map = mallorca + pluviometros
bars_overlay = bars.encode(color=alt.value("red")).transform_filter(brush)
right bars = alt. layer(bars, bars_overlay)
left_map | (bars + right_bars)

為簡(jiǎn)潔起見，下面僅使用 2013 年的數(shù)據(jù)為每種車輛類型生成一張圖表。

import altair as alt
# Base chart
url_geojson = 'mallorca_geoson.json'
data_geojson = alt. Data(url=url_geojson, format=alt.DataFormat(property='features',type='json'))
mallorca = alt.Chart(data_geojson), mark_geoshape(stroke="gray", strokeWidth=0.2)data = data. loc[data. year == 2013, :]options = ["cars", "scooters", "motorbikes", "vans", "trucks"]
mallorca \.transform_lookup(lookup="properties, neighbourhood",from_alt.LookupData(data=data, key="municipality", fields=["municipality"] + vehic)) \.encode(alt.Color(alt, repeat('row'), type='quantitative'),tooltip=["municipality:N", alt.Tooltip(alt.repeat('row'), type='quantitative')]) \.project(type="equalEarth") \.properties(width=400, height=200) \.repeat(row=options) \.resolve_scale(color='independent')

Plotly 允許以與 Altair 提供的類似交互的方式表示地圖，此外還可以通過 mapbox API 訪問 Carto 和 OpenStreet 地圖。

import plotly.express as px
fig = px.scatter_mapbox(pluvio,lat=pluvio. COORD Y,lon=pluvio.COORD_X,hover_name="NOMBRE",hover data="ALTITUD",mapbox_style="carto-positron",center={"lat": 39.5, "lon": 2.85},zoom=8)
fig.show()

下面的交互式地圖與 Dash 集成，包含額外的交互，可讓您探索多年來不同車輛的數(shù)量。

import json
from dash import Dash, dcc, html, Input, Output
with open('mallorca_geoson.json') as file:mallorca_geoson = json.load(file)
app = Dash (__name__)
app.layout = html.Div([html.H4('Vehicles per 1000 inhab.'),html.P("Select a vehicle:"),dcc.RadioItems(id='mapbox-mallorca geoson-choropleth-x-vehicle',options=["cars", "scooters", "motorbikes", "vans", "trucks"],value="cars",inline=True),dcc.Graph(id="mapbox-mallorca geoson-choropleth-x-graph"),
])
@app.callback(Output("mapbox-mallorca_geoson-choropleth-x-graph", "figure"),Input("mapbox-mallorca_geoson-choropleth-x-vehicle", "value"))
def display choropleth(vehicle):fig = px.choropleth_mapbox(data, geojson=mallorca_geoson, color=vehicle,color continuous scale="Viridis",locatīons="municípality",featureidkey="properties.neighbourhood",center={"lat": 39.5, "lon": 2.85},zoom=7.5,animation_frame="year")fig.update_layout(margin={"r":0,"tt":\theta,"l":0,"b":0},mapbox_accesstoken=token)return fig
if __name___= "__main__":app.run_server (debug=True)

Bokeh 是一個(gè)用于創(chuàng)建交互式 JS 可視化的 Python 庫，它不基于 D3.js。如果 Plotly 與包含 Flask Web 服務(wù)器的 Dash 集成，那么 Bokeh 則使用 Tornado Web 服務(wù)器，后者在后端使用 WebSockets。WebSockets 是有狀態(tài)的且異步的。

from bokeh.io import output_file, show, output_notebook
from bokeh.layouts import column
from bokeh.models import WMTSTileSource, ColumnDataSource, LinearColorMapper, ColorBar, Select, I
HoverTool, PrintfTickFormatter
from bokeh.models.widgets import RadioButtonGroup
from bokeh.palettes import Viridis6
from bokeh.plotting import figure, show
from numpy import pi, tan, log
import json
output notebook()

def transform_to_mercator(data, lat, lon):k=637813data["x"] = data[lon] * k * pi / 180.data["y"] = log(tan((90 + data[lat]) * p1 / 360.)) * kreturn data

bokeh_pluvio = transform_to_mercator(pluvio, "COORD_Y", "COORD_X")
x_range, y_range = ((250000, 390000), (4750000, 4850000))
tooltips = [("Name", "@NOMBRE"), ("Altitud", "@ALTITUD"), ("(Lon, Lat)", "($x, $y)")]
source = ColumnDataSource(data=bokeh_pluvio)
hover = HoverTool(tooltips=tooltips)
p = figure(tools="pan, wheel_zoom, hover, reset",x_range =x_range,y_range=y_range,x_axis_type="mercator",y_axis_type="mercator",
)p.circle(source=source, fill_color="blue", size=10)
url="http://a.basemaps.cartocdn.com/rastertiles/voyager/{Z}/{X}/{Y}.png"
attribution = "Credit: Carto, under CC BY 3.0. Data by OSM, under ODbL"
p.add_tile(WMTSTileSource(url=url, attribution=attribution))
show(p)

👉參閱、更新：計(jì)算思維 | 亞圖跨際