7. Chlorophyll estimate trends

This notebook is optimized to run in Google Colab.

In [ ]:

!pip install fast-dash

!pip install fast-dash

In [ ]:

import datetime
import pandas as pd
import plotly.express as px

from fast_dash import fastdash, Graph

import datetime import pandas as pd import plotly.express as px from fast_dash import fastdash, Graph

In [ ]:

# If you get a ConnectionError, please retry. 
data = pd.read_csv("https://raw.githubusercontent.com/dkedar7/fast_dash/docs/docs/Examples/assets/chla_subset.csv")
data['date_acquired'] = pd.to_datetime(data["date_acquired"])
data = data.sort_values(["date_acquired", "gnis_name"])
data

# If you get a ConnectionError, please retry. data = pd.read_csv("https://raw.githubusercontent.com/dkedar7/fast_dash/docs/docs/Examples/assets/chla_subset.csv") data['date_acquired'] = pd.to_datetime(data["date_acquired"]) data = data.sort_values(["date_acquired", "gnis_name"]) data

In [ ]:

Let's build a simple function to display trends¶

In [ ]:

#@fastdash(mode="external", port=5000)
def plot_chlorophyll_estimates(lake_name: str = data.gnis_name.unique().tolist()) -> Graph:
    """
    Plot Landsat-driven estimation of Chlorophyll-a values in island waters of NY state.
    """
    
    filtered_data = data[data["gnis_name"] == lake_name]
    trend = px.line(data_frame=filtered_data, x="date_acquired", y="predictions", template="simple_white")
    
    return trend

#@fastdash(mode="external", port=5000) def plot_chlorophyll_estimates(lake_name: str = data.gnis_name.unique().tolist()) -> Graph: """ Plot Landsat-driven estimation of Chlorophyll-a values in island waters of NY state. """ filtered_data = data[data["gnis_name"] == lake_name] trend = px.line(data_frame=filtered_data, x="date_acquired", y="predictions", template="simple_white") return trend

In [ ]:

plot_chlorophyll_estimates("Cayuta Lake")

plot_chlorophyll_estimates("Cayuta Lake")

In [ ]:

Display mean and median values¶

In [ ]:

# Define the layout of the app. Alphabets denote the placement of each output
# In this layout, we are telling the app to display the first two outputs (A and B) next to each other,
# and let the third output occupy double the width in the second row.
output_layout = """
AB
CC
CC
"""

@fastdash(mode="external", port=5000, mosaic=output_layout)
def plot_chlorophyll_estimates(lake_name: str = data.gnis_name.unique().tolist()) -> (str, str, Graph):
    """
    Plot Landsat-driven estimation of Chlorophyll-a values in island waters of NY state.
    """
    
    filtered_data = data[data["gnis_name"] == lake_name]
    trend = px.line(data_frame=filtered_data, x="date_acquired", y="predictions", template="simple_white")
    
    mean = f"{round(filtered_data['predictions'].mean(), 2)} ug/L"
    median = f"{round(filtered_data['predictions'].median(), 2)} ug/L"
    
    return mean, median, trend

# Define the layout of the app. Alphabets denote the placement of each output # In this layout, we are telling the app to display the first two outputs (A and B) next to each other, # and let the third output occupy double the width in the second row. output_layout = """ AB CC CC """ @fastdash(mode="external", port=5000, mosaic=output_layout) def plot_chlorophyll_estimates(lake_name: str = data.gnis_name.unique().tolist()) -> (str, str, Graph): """ Plot Landsat-driven estimation of Chlorophyll-a values in island waters of NY state. """ filtered_data = data[data["gnis_name"] == lake_name] trend = px.line(data_frame=filtered_data, x="date_acquired", y="predictions", template="simple_white") mean = f"{round(filtered_data['predictions'].mean(), 2)} ug/L" median = f"{round(filtered_data['predictions'].median(), 2)} ug/L" return mean, median, trend

In [ ]:

Add date filters¶

In [ ]:

# Define the layout of the app. Alphabets denote the placement of each output
# In this layout, we are telling the app to display the first two outputs (A and B) next to each other,
# and let the third output occupy double the width in the second row.
output_layout = """
AB
CC
CC
"""

@fastdash(mode="external", port=5000, mosaic=output_layout)
def plot_chlorophyll_estimates(lake_name: str = data.gnis_name.unique().tolist(), 
                               start_date=datetime.date(2015, 1, 1),
                              end_date=datetime.date(2023, 12, 31)) -> (str, str, Graph):
    """
    Plot Landsat-driven estimation of Chlorophyll-a values in island waters of NY state.
    """
    
    filtered_data = data[(data["gnis_name"] == lake_name) & (data["date_acquired"].between(start_date, end_date))]
    trend = px.line(data_frame=filtered_data, x="date_acquired", y="predictions", template="simple_white")
    
    mean = f"{round(filtered_data['predictions'].mean(), 2)} ug/L"
    median = f"{round(filtered_data['predictions'].median(), 2)} ug/L"
    
    return mean, median, trend

# Define the layout of the app. Alphabets denote the placement of each output # In this layout, we are telling the app to display the first two outputs (A and B) next to each other, # and let the third output occupy double the width in the second row. output_layout = """ AB CC CC """ @fastdash(mode="external", port=5000, mosaic=output_layout) def plot_chlorophyll_estimates(lake_name: str = data.gnis_name.unique().tolist(), start_date=datetime.date(2015, 1, 1), end_date=datetime.date(2023, 12, 31)) -> (str, str, Graph): """ Plot Landsat-driven estimation of Chlorophyll-a values in island waters of NY state. """ filtered_data = data[(data["gnis_name"] == lake_name) & (data["date_acquired"].between(start_date, end_date))] trend = px.line(data_frame=filtered_data, x="date_acquired", y="predictions", template="simple_white") mean = f"{round(filtered_data['predictions'].mean(), 2)} ug/L" median = f"{round(filtered_data['predictions'].median(), 2)} ug/L" return mean, median, trend

In [ ]:

How do we draw a map?¶

In [ ]:

!pip install folium

!pip install folium

In [ ]:

import folium
from fast_dash import html

# Define the layout of the app. Alphabets denote the placement of each output
# In this layout, we are telling the app to display the first two outputs (A and B) next to each other,
# and let the third output occupy double the width in the second row.
output_layout = """
DDAB
CCCC
CCCC
"""

@fastdash(mode="external", port=5000, mosaic=output_layout)
def plot_chlorophyll_estimates(lake_name: str = data.gnis_name.unique().tolist(), 
                               start_date=datetime.date(2015, 1, 1),
                              end_date=datetime.date(2023, 12, 31)) -> (str, str, Graph, html.Iframe()):
    """
    Plot Landsat-driven estimation of Chlorophyll-a values in island waters of NY state.
    """
    
    filtered_data = data[(data["gnis_name"] == lake_name) & (data["date_acquired"].between(start_date, end_date))]

    centroid_latitude = filtered_data["centroid_latitude"].dropna().iloc[0]
    centroid_longitude = filtered_data["centroid_longitude"].dropna().iloc[0]
    
    map_ = folium.Map(location=(centroid_latitude, centroid_longitude), zoom_start=13)._repr_html_()
    trend = px.line(data_frame=filtered_data, x="date_acquired", y="predictions", template="simple_white")
    
    mean = f"{round(filtered_data['predictions'].mean(), 2)} ug/L"
    median = f"{round(filtered_data['predictions'].median(), 2)} ug/L"
    
    return mean, median, trend, map_

import folium from fast_dash import html # Define the layout of the app. Alphabets denote the placement of each output # In this layout, we are telling the app to display the first two outputs (A and B) next to each other, # and let the third output occupy double the width in the second row. output_layout = """ DDAB CCCC CCCC """ @fastdash(mode="external", port=5000, mosaic=output_layout) def plot_chlorophyll_estimates(lake_name: str = data.gnis_name.unique().tolist(), start_date=datetime.date(2015, 1, 1), end_date=datetime.date(2023, 12, 31)) -> (str, str, Graph, html.Iframe()): """ Plot Landsat-driven estimation of Chlorophyll-a values in island waters of NY state. """ filtered_data = data[(data["gnis_name"] == lake_name) & (data["date_acquired"].between(start_date, end_date))] centroid_latitude = filtered_data["centroid_latitude"].dropna().iloc[0] centroid_longitude = filtered_data["centroid_longitude"].dropna().iloc[0] map_ = folium.Map(location=(centroid_latitude, centroid_longitude), zoom_start=13)._repr_html_() trend = px.line(data_frame=filtered_data, x="date_acquired", y="predictions", template="simple_white") mean = f"{round(filtered_data['predictions'].mean(), 2)} ug/L" median = f"{round(filtered_data['predictions'].median(), 2)} ug/L" return mean, median, trend, map_

In [ ]: