Weather Anomaly Detection & Notification System

This project fetches weather forecast data, detects specific weather anomalies, and sends WhatsApp notifications to subscribed users via the Zawa API. It is designed for localized weather monitoring (example: Krandegan) and uses BMKG forecast data.

---

🚀 Features

• Automatic hourly weather fetching from database.

• Category-based anomaly detection:

  • Heavy Rain (Hujan Lebat) → High cloud coverage beyond threshold and weather description contains berawan, kabut, or hujan.
  • Prolonged Drought (Kemarau Panjang) → Humidity below threshold and description contains cerah.

• Real-time WhatsApp notifications using the Zawa API.

• Message aggregation → Multiple anomalies in one day are combined into a single message to avoid spam.

• MongoDB upsert support → New data is inserted, old records are updated automatically.

• .env support → Phone numbers and API credentials are stored securely.

---

📂 Project Structure

.
├── server/
│   ├── api/                  # API routes (QR connect, authorization, etc.)
│   ├── data/                 # Weather anomaly detection & notification logic
│   └── index.js              # App entry point
├── .env                      # Environment variables (not committed to GitHub)
├── package.json
└── README.md

---

⚙️ How It Works

1. Data Fetching The system queries the latest summarized weather data from MongoDB every hour. It also receives updates to past data and uses MongoDB's updateOne(..., { upsert: true }) to insert or update records.

2. Anomaly Detection The detection logic is category-based:

   • Heavy Rain:

     tcc > cloudThreshold && /(berawan|kabut|hujan)/i.test(description)

   • Prolonged Drought:

     hu < humidityThreshold && /cerah/i.test(description)

3. Notification If anomalies match today’s date, a WhatsApp message is prepared:

   Peringatan Cuaca Krandegan (BMKG):
   
   Pada Senin, 11 Agustus 2025 pukul 15:00, terdeteksi kategori *Hujan Lebat* (Hujan Ringan) dengan tutupan awan 85% dan kelembapan 92%.
   
   Info lebih lengkap: https://www.bmkg.go.id/cuaca/potensi-cuaca-ekstrem
   

   If no anomalies are detected:

   Hari ini tidak ada potensi hujan lebat atau kemarau panjang yang terdeteksi.
   

4. Message Delivery Uses the Zawa API to send messages to phone numbers stored in .env.

---

🔑 Environment Variables

Create a .env file in the root directory:

DB_USER=mongo_user
DB_PASS=mongo_password
DB_URL=mongo_db_uri
ID=ZAWA_ID
SESS_ID=ZAWA_SESS_ID
PHONES=list_of_subscribers_phone_numbers

---

📡 API Endpoints

GET /api/daftar

Authorizes and registers the current Zawa session.

GET /api/hubungkan

Fetches a QR code (base64) from Zawa API and returns it as an image.

GET /api/ulang

Retries the registration process

GET /api/hapus

Deletes the current session

More info at Zawa's

GET /data/weather

Retrieves overall weather data

GET /data/anomaly

Retrieves anomalous weather

---

🛠 Installation

git clone https://github.com/lindduncoding/early-warning-kkn.git
cd weather-anomaly-notifier
npm install

---

▶️ Running the Project

node server/index.js

# or for always on deployment
nohup node server/index.js > output.log 2>&1 &

The app will:

1. Fetch the latest weather data.
2. Detect anomalies.
3. Send WhatsApp notifications.

It does this every 8 hours

---

🧪 Example Output

With Anomaly

Peringatan Cuaca Krandegan (BMKG):

Pada Senin, 11 Agustus 2025 pukul 15:00, terdeteksi kategori *Hujan Lebat* (Hujan Ringan) dengan tutupan awan 85% dan kelembapan 92%.

Info lebih lengkap: https://www.bmkg.go.id/cuaca/potensi-cuaca-ekstrem

Without Anomaly

Hari ini tidak ada potensi hujan lebat atau kemarau panjang yang terdeteksi.

---