Water Quality & Hypoxia

Gulf of Mexico Dead Zone (USA)

Scale: 5,000–22,000 km² annually (largest in Western Hemisphere) | Driver: Mississippi–Atchafalaya River N/P runoff (~1.6 Mt N/yr) | Impact: ~$1 billion annual loss to fisheries & tourism

The Gulf hypoxia forms each summer in response to peak freshwater discharge and nutrient loading. Seasonal breakdown occurs in autumn. Long-term observations show expansion over decades, driven by increasing agricultural intensification in the Mississippi basin. Mitigation requires coordinated multi-state efforts.

Baltic Sea Hypoxia (Northern Europe)

Scale: ~60,000 km² seasonal anoxic/hypoxic bottom waters | Driver: Riverine N/P input + limited water exchange with Atlantic | Impact: Benthic biodiversity collapse, fisheries stress

The Baltic represents the world's largest artificial dead zone by area. Its restricted basins and cyclonic wind patterns limit oxygen replenishment. Recent years show increased anoxic events, with some studies projecting further expansion under climate change scenarios.

Chesapeake Bay (USA)

Scale: ~2,000–5,000 km² hypoxic habitat annually | Driver: Potomac, Susquehanna, James Rivers; urban & agricultural runoff | Impact: Blue crab migration, oyster and fish kills

Chesapeake Bay has been a laboratory for hypoxia management since the 1980s. Ongoing nutrient load reductions show promise, yet recovery is slow. The bay demonstrates the challenge of addressing legacy sediment loads and the lag between land-use change and water quality improvement.

Yangtze River Estuary (China)

Scale: ~14,000 km² hypoxic area (growing) | Driver: Massive N/P loading from agricultural & industrial sources | Impact: Ecosystem shifts, economic losses to aquaculture

East Asian systems are experiencing rapid eutrophication due to intensive agriculture and urbanization. The Yangtze estuary's hypoxic zone has expanded significantly over the past two decades, raising concerns for regional food security.

# Import Required Libraries
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats

# Set visual style
sns.set_style('whitegrid')
plt.rcParams['figure.figsize'] = (12, 6)

# Generate synthetic 3-year water quality dataset
np.random.seed(42)
n_days = 365 * 3
dates = pd.date_range(start='2022-01-01', periods=n_days, freq='D')
day_of_year = np.array([(d.timetuple().tm_yday) for d in dates])

# Dissolved Oxygen dynamics (mg/L)
do_surface = 8 + 2.5*np.cos(2*np.pi*day_of_year/365) + np.random.normal(0, 0.3, n_days)
do_bottom = 6 + 4*np.cos(2*np.pi*day_of_year/365) + np.random.normal(0, 0.4, n_days)
do_bottom = np.clip(do_bottom, 0, 10)

# Create DataFrame with water quality variables
df = pd.DataFrame({
    'date': dates,
    'do_surface': do_surface,
    'do_bottom': do_bottom,
    'chlorophyll_a': 2 + 1.5*np.sin(2*np.pi*day_of_year/365) + np.random.normal(0, 0.1, n_days),
    'nitrate': 20 + 15*np.cos(2*np.pi*day_of_year/365) + np.random.normal(0, 2, n_days),
    'temperature': 18 + 10*np.sin(2*np.pi*day_of_year/365) + np.random.normal(0, 0.5, n_days)
})

# Define hypoxia threshold (DO < 2 mg/L)
df['hypoxic'] = df['do_bottom'] < 2

# Monthly summary statistics
df['month'] = df['date'].dt.month
monthly_avg = df.groupby('month')[['do_surface', 'do_bottom', 'chlorophyll_a']].mean()

print('Hypoxia Statistics:')
print(f'Total hypoxic days: {df["hypoxic"].sum()} of {len(df)} days')
print(f'Hypoxia frequency: {df["hypoxic"].sum()/len(df)*100:.1f}%')

# Correlation between nutrients and bottom DO
correlation = df['nitrate'].corr(df['do_bottom'])
print(f'Nitrate-DO Correlation: {correlation:.3f}')

# Plot seasonal DO cycle
fig, ax = plt.subplots(figsize=(12, 6))
months = ['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep','Oct','Nov','Dec']
ax.plot(monthly_avg.index, monthly_avg['do_surface'], 'o-', label='Surface DO', linewidth=2)
ax.plot(monthly_avg.index, monthly_avg['do_bottom'], 's-', label='Bottom DO', linewidth=2)
ax.axhline(y=2, color='red', linestyle='--', linewidth=2, label='Hypoxia Threshold (2 mg/L)')
ax.fill_between(monthly_avg.index, 0, 2, alpha=0.2, color='red')
ax.set_xlabel('Month')
ax.set_ylabel('Dissolved Oxygen (mg/L)')
ax.set_title('Seasonal Dissolved Oxygen Cycle in a Stratified Estuary')
ax.set_xticks(range(1, 13))
ax.set_xticklabels(months)
ax.legend()
ax.grid(True, alpha=0.3)
plt.show()

# Vertical DO profile analysis
depths = np.linspace(0, 30, 30)

# Summer profile: Strong stratification with hypoxia
do_summer = 7.5 - 0.2*depths - 0.8*(np.maximum(0, depths-8)/6)**2
temp_summer = 28 - 0.3*depths - 1.2*(np.maximum(0, depths-8)/6)**2

# Winter profile: Well-mixed, oxygen-rich
do_winter = 9.5 - 0.1*depths
temp_winter = 12 - 0.05*depths

fig, axes = plt.subplots(1, 2, figsize=(12, 6))

# Summer profile
ax1 = axes[0]
ax1.plot(do_summer, -depths, 'o-', label='DO (mg/L)', linewidth=2, color='#e74c3c')
ax1.axvline(x=2, color='orange', linestyle='--', linewidth=2, label='Hypoxia Threshold')
ax1.fill_betweenx(-depths, 0, 2, alpha=0.15, color='red')
ax1.set_xlabel('Dissolved Oxygen (mg/L)')
ax1.set_ylabel('Depth (m)')
ax1.set_title('Summer: Strong Stratification with Hypoxia')
ax1.invert_yaxis()
ax1.grid(True, alpha=0.3)
ax1.legend()

# Winter profile
ax2 = axes[1]
ax2.plot(do_winter, -depths, 'o-', label='DO (mg/L)', linewidth=2, color='#2ecc71')
ax2.axvline(x=2, color='orange', linestyle='--', linewidth=2, label='Hypoxia Threshold')
ax2.set_xlabel('Dissolved Oxygen (mg/L)')
ax2.set_ylabel('Depth (m)')
ax2.set_title('Winter: Well-Mixed, No Hypoxia')
ax2.invert_yaxis()
ax2.grid(True, alpha=0.3)
ax2.legend()

plt.tight_layout()
plt.show()

# Lagged correlation: Nutrients leading to DO depletion
lags = range(0, 31)
lag_correlations = []
for lag in lags:
    if lag == 0:
        corr = df['nitrate'].corr(df['do_bottom'])
    else:
        corr = df['nitrate'].iloc[:-lag].corr(df['do_bottom'].iloc[lag:])
    lag_correlations.append(corr)

fig, ax = plt.subplots(figsize=(10, 6))
ax.plot(lags, lag_correlations, 'o-', linewidth=2, markersize=6)
ax.axhline(y=0, color='black', linestyle='-', linewidth=1)
ax.set_xlabel('Lag (days)')
ax.set_ylabel('Correlation Coefficient')
ax.set_title('Lagged Correlation: Nitrate → Bottom DO Depletion')
ax.grid(True, alpha=0.3)
plt.show()

print('\nAnalysis complete! Export plots as PNG or PDF for reports.')

How to Use This Code:

• Environment: Run in Jupyter Notebook, JupyterLab, or VS Code with Python extension
• Dependencies: Install with pip install numpy pandas matplotlib seaborn scipy
• Data: Replace the synthetic dataset generation with your own water quality CSV file
• Customization: Adjust hypoxia threshold, depth ranges, and month labels to match your study region
• Output: Saves figures as high-resolution PNG; modify with plt.savefig('filename.png', dpi=300)

💡 Pro Tip: For real data, load CSV files using df = pd.read_csv('your_data.csv', parse_dates=['date']) and ensure date columns are datetime objects for time series operations.