Preswald | Blog

Tracking and analyzing earthquake data from CSV files was a time-consuming, manual process. Our team relied on multiple tools to clean, filter, and visualize seismic activity data, making real-time insights difficult to generate.

We needed a way to automate earthquake analytics, eliminate repetitive tasks, and build interactive dashboards to help researchers, emergency responders, and data analysts gain real-time insights from historical earthquake records.

That’s when we turned to Preswald.

The Challenge: Slow and Manual Earthquake Data Processing

Our earthquake analytics workflow had several bottlenecks:

🌍 Data Spread Across Multiple CSV Files – Downloading and merging data from various sources was tedious.
🧼 Manual Cleaning and Filtering – Formatting inconsistencies meant constant adjustments.
📊 Basic Static Reports – We relied on Excel and static charts, which lacked interactivity.
😩 Inefficient Scripting – Custom Python scripts were hard to maintain and deploy.

We needed a fast, automated, and interactive way to:

Load and process CSV earthquake data in real-time.
Filter and visualize earthquakes by magnitude, depth, and location.
Deploy an interactive web dashboard without needing a frontend.

Why We Chose Preswald

Preswald provided an all-in-one solution to automate earthquake analytics:

✅ Instant CSV data ingestion – No manual importing or cleaning.
✅ Interactive dashboards with Python – No need for JavaScript or BI tools.
✅ Automated filtering and visualization – Earthquake trends in real-time.
✅ One-command deployment – Shareable dashboards in minutes.

The Implementation: From Static Reports to Live Dashboards

Step 1: Installing Preswald

We started by installing Preswald:

pip install preswald

Step 2: Setting Up the Earthquake Analytics App

We created a new Preswald project:

preswald init earthquake_dashboard
cd earthquake_dashboard

This automatically generated a pre-configured structure, including:

hello.py – The main earthquake analytics script.
preswald.toml – Configuration for themes and settings.
secrets.toml – Secure storage for sensitive credentials.

Step 3: Loading and Connecting to Earthquake Data (CSV)

Instead of manually loading data every time, we automated CSV ingestion:

import pandas as pd
from preswald import connect, get_df, text, table, slider, plotly
import plotly.express as px

# Title
text("# Earthquake Analytics Dashboard 🌍")

# Load earthquake data from CSV
connect()
data = get_df("earthquake_data")

# Display raw data
table(data)

✅ No more manual file uploads – Preswald reads CSVs dynamically.

Step 4: Filtering Earthquake Data by Magnitude

To allow users to filter earthquakes based on severity, we added a magnitude slider:

# Slider for filtering magnitude
min_magnitude = slider("Minimum Magnitude", min_val=0.0, max_val=10.0, default=5.0)

# Convert Magnitude column to numeric and filter data
data["Magnitude"] = pd.to_numeric(data["Magnitude"], errors="coerce")
filtered_data = data[data["Magnitude"] >= min_magnitude]

# Show filtered earthquakes
table(filtered_data)

# Summary statistics
text(f"### Total Earthquakes with Magnitude ≥ {min_magnitude}: {len(filtered_data)}")

✅ Instant filtering – Users can adjust the magnitude threshold dynamically.

Step 5: Visualizing Earthquake Locations on an Interactive Map

Using Plotly, we generated an interactive map of earthquake locations:

# Interactive map using Plotly
text("## Earthquake Locations")
fig_map = px.scatter_geo(
    filtered_data,
    lat="Latitude",
    lon="Longitude",
    color="Magnitude",
    size="Magnitude",
    hover_name="ID",
    title="Earthquake Map",
)
plotly(fig_map)

✅ No complex GIS software needed – The map updates dynamically based on selected filters.

Step 6: Analyzing Earthquake Magnitudes and Depths

We added two key visualizations to help users analyze magnitude distribution and depth impact:

# Magnitude distribution histogram
fig_hist = px.histogram(
    filtered_data, x="Magnitude", nbins=20, title="Distribution of Magnitudes"
)
plotly(fig_hist)

# Depth vs. Magnitude scatter plot
fig_scatter = px.scatter(
    filtered_data,
    x="Depth",
    y="Magnitude",
    color="Magnitude",
    title="Depth vs Magnitude",
    labels={"Depth": "Depth (km)", "Magnitude": "Magnitude"},
)
plotly(fig_scatter)

✅ Clear trends – Users can spot patterns between depth and earthquake severity.

Step 7: Running the Dashboard Locally

With just one command, our dashboard was live:

preswald run

📍 The app was instantly available at:

🌐 http://localhost:8501

Step 8: Deploying to the Cloud for Public Access

To make the dashboard accessible to researchers worldwide, we deployed it to the cloud:

preswald deploy --target structured

✅ Anyone can now interact with real-time earthquake data without needing Python installed.

The Results: Real-Time, Automated Earthquake Analysis

🚀 Transformed static CSV files into an interactive web dashboard
🚀 Eliminated manual filtering and visualization work
🚀 Enabled real-time earthquake tracking with dynamic controls
🚀 Deployed a shareable cloud dashboard in minutes

Final Thoughts: Preswald for AI-Driven Data Analytics

If your team struggles with analyzing large CSV datasets manually, Preswald is a game-changer. It allows you to build interactive, AI-powered dashboards in Python without complex frontend coding.

Ready to revolutionize your data analytics?

🚀 Try Preswald today!

🔗 https://github.com/StructuredLabs/preswald

How we visualized earthquake data using Preswald