Chapter 3 Practical 4

Time-Series Analysis of Natural Resource Consumption

PART A

1)  Natural Resource analyzer

Aim:
To acquire, organize, and analyze time-series data on the consumption of a key natural resource (energy, water, minerals) for a selected country using publicly available data and to identify trends, patterns, and driving factors using the analytical tools available in Microsoft Excel.

Principle:
Time-series analysis involves studying data points collected or recorded at specific time intervals. By analyzing this data, we can:

1. Identify Trends: Long-term increases or decreases in consumption.

2. Understand Seasonality: Regular, predictable patterns that repeat over a specific period (e.g., yearly cycles).

3. Forecast: Make informed predictions about future consumption based on historical patterns.
   This is crucial for environmental science as it helps understand the impact of economic growth, policy changes, and technological shifts on resource use and environmental pressure.

Materials Required:

1. Computer: With internet access.

2. Microsoft Excel: Installed on the computer (version 2016 or later is ideal for access to all used features).

3. Data Sources:

   IEA (https://www.iea.org/data-and-statistics/data-tools/energy-statistics-data-browser?country=INDIA&energy=Balances&year=2017)  

   • World Bank Open Data: https://data.worldbank.org/country/india (Search for "Energy use", "Freshwater withdrawal", "Forest area")

   • BP Statistical Review of World Energy: https://www.bp.com/en/global/corporate/energy-economics/statistical-review-of-world-energy.html (Excellent for fossil fuel, renewable energy data)

   • UN Data: http://data.un.org/

   • National Statistical Offices: (e.g., US EIA, China's NBS)

     Regression Calculator (https://www.pearson.com/channels/calculators/linear-regression-calculator) 

     
     

Procedure:

Phase 1: Data Selection and Acquisition (1-2 Hours)

1. Select a Country and Resource: Choose a country and a natural resource.

   • Example:

     • Country: United States

     • Resource: Natural Gas Consumption

     • Rationale: The US is a top consumer, and natural gas is a key transition fuel.

2. Acquire the Data:

   • Go to the BP Statistical Review/ IEA website.

   • Download the Excel spreadsheet for the latest year.

   • Open the file and find the sheet for "Natural Gas Consumption".

   • Locate the data for your chosen country. The data is typically provided in billion cubic meters(TJ for IEA).

   • Select the time series data (e.g., from 1990 to 2023). Include the years in one column and the values in the next column.

   • Copy this data into a new, clean Excel workbook.

      

      

Phase 2: Data Preparation and Cleaning in Excel (1 Hour)

1. Set Up Your Sheet:

   • Column A: Year

   • Column B: Natural Gas Consumption (billion cubic meters)

   • Ensure there are no missing years. If data is missing, you can leave the cell blank, but a continuous series is best.

2. Calculate Additional Metrics (Important for Analysis):

   • Column C: Annual Change. In cell C3, enter the formula =(B3-B2)/B2. Format as a percentage. Drag this formula down to the end of your data. This shows the year-on-year growth rate.

   • Column D: 5-Year Moving Average. This smooths out short-term fluctuations and reveals the long-term trend.

     • In cell D6, enter the formula =AVERAGE(B2:B6).

     • Drag this formula down. You will have mean value... for the first four years.

Phase 3: Data Visualization and Trend Analysis (2 Hours)

1. Create a Line Chart:

   • Select columns A, B, and D (Year, Consumption, Moving Average).

   • Go to Insert > Charts > Line Chart.

   • This will create a chart showing the raw data and the smoothed trend line.

2. Add Chart Elements:

   • Add a Chart Title (e.g., "Natural Gas Consumption in the United States (1990-2023)").

   • Add Axis Titles (e.g., "Year" for X-axis, "Billion Cubic Meters" for Y-axis).

   • Add a Legend to distinguish the lines.

3. Add a Linear Trendline:

   • Click on the "Natural Gas Consumption" data series in the chart.

   • Right-click and select "Add Trendline".

   • In the Format Trendline pane, choose Linear.

   • Check the boxes for "Display Equation on chart" and "Display R-squared value on chart".

   • The equation (y = mx + b) shows the average annual change (slope, m).

   • The R-squared value (R²) indicates how well the trendline fits the data (closer to 1 = better fit).

      

      

      

Phase 4: Advanced Analysis with Excel's Data Analysis Toolpak (1 Hour)
Note: You may need to enable the Data Analysis Toolpak in Excel Options > Add-ins.

1. Calculate Descriptive Statistics:

   • Go to Data > Data Analysis > Descriptive Statistics.

   • Select your input range (Column B: Consumption data).

   • Check "Summary statistics". This will generate a table showing mean, median, standard deviation, etc. This tells you about the central tendency and variability of consumption.

2. Perform a Regression Analysis (Optional but valuable):

   • This helps quantify the relationship between time and consumption.

   • In a new column, create a "Time Index" (1, 2, 3, ... for each year).

   • Go to Data > Data Analysis > Regression.

   • Input Y Range: Your consumption data (B:B).

   • Input X Range: Your Time Index column.

   • Click OK. Excel will output a regression analysis sheet. The "Coefficients" for the X Variable give you the slope of the trendline (the annual change).

Phase 5: Interpretation and Report Writing (2 Hours)
Synthesize your findings into a short report.

---

Observations & Data Analysis:

Table 1: Descriptive Statistics for US Natural Gas Consumption (1990-2023)

Statistic	Value
Mean	[Value from Excel] billion cubic meters
Standard Deviation	[Value from Excel]
Minimum	[Value from Excel] (Year: [Year])
Maximum	[Value from Excel] (Year: [Year])

Chart 1: Time-Series Plot of US Natural Gas Consumption
(Students will paste their Excel chart here)

• Chart Title: Natural Gas Consumption in the United States (1990-2023)

• Trendline Equation: y = [slope]x + [intercept]

• R² Value: [value]

Interpretation of Chart:

• Trend: The trendline has a positive slope of [slope], indicating an overall increasing trend in natural gas consumption over the 33-year period.

• Variability: The data shows significant variability around the trend. Notable periods of decline are visible around [Year] (likely due to the 2008 financial crisis) and [Year] (likely due to COVID-19 pandemic impacts).

• Recent Changes: Consumption appears to have plateaued or slightly declined in the most recent years ([Years]), potentially indicating market saturation or a shift towards other energy sources.

---

Discussion:

1. Linking Trends to Real-World Events:

   • The overall increase aligns with the US shale gas revolution (post-2008), which dramatically increased supply and lowered prices, leading to higher consumption, especially for electricity generation (displacing coal).

   • The sharp declines correlate with economic recessions (2001, 2008) and the COVID-19 pandemic (2020), where reduced industrial activity led to a drop in energy demand.

2. Environmental Implications:

   • While natural gas burns cleaner than coal (less CO₂ per unit of energy), this sustained increase in consumption still represents a significant source of greenhouse gas emissions (CO₂ and methane leaks).

   • The trend highlights the challenge of climate change mitigation: even as the fuel mix gets cleaner, absolute consumption levels can still rise, potentially negating gains from switching away from dirtier fuels.

3. Limitations of the Analysis:

   • Correlation vs. Causation: Our analysis shows a relationship between time and consumption but doesn't prove what caused the changes. Deeper analysis would require adding more variables (e.g., GDP, population, gas prices).

   • Excel's Limitations: Excel is excellent for visualization and basic analysis but lacks more sophisticated time-series tools found in R or Python (e.g., ARIMA modeling for forecasting). The linear trendline is a simplification of a potentially more complex trend.

   • Data Quality: The analysis is only as good as the underlying data. Public data can sometimes have estimation errors or methodological changes over time.

Conclusion:

This practical demonstrated that Microsoft Excel is a powerful and accessible tool for conducting preliminary time-series analysis of environmental data. The analysis of US natural gas consumption from 1990 to 2023 revealed a clear upward trend, heavily influenced by technological breakthroughs (fracking) and macroeconomic shocks. The R² value of [value] indicates that the linear trend explains a significant portion of the variation in consumption over time. This type of analysis forms the foundational step for understanding resource dynamics, informing policy decisions, and critically evaluating the progress (or lack thereof) towards sustainability goals. It underscores the fact that environmental science is increasingly a data-driven field.

Viva Voce Questions:

1. What does the slope of your linear trendline represent in real-world terms?
   The slope (the 'm' value in y = mx + b) represents the average annual increase in natural gas consumption, in billion cubic meters, over the entire time period studied. For example, a slope of 5 would mean consumption grew by an average of 5 billion cubic meters each year.

2. Why did we calculate a moving average, and what does it show that the raw data does not?
   The moving average smooths out short-term volatility and noise (e.g., year-to-year fluctuations caused by unusual weather or minor economic shifts). This makes the underlying long-term trend much clearer and easier to visualize and interpret.

3. If your R-squared value was very low (e.g., 0.15), what would that tell you about the trend?
   A low R-squared value would indicate that the linear trendline is a very poor fit for the data. It would suggest that there is no strong, consistent upward or downward trend, and that changes in consumption are driven more by random variation or complex, non-linear factors that a straight line cannot capture.

4. Besides economic recessions, what other factors could cause a sudden dip in a natural resource consumption time series?

   • Technological Shifts: A rapid adoption of a more efficient technology.

   • Policy Changes: The introduction of a carbon tax or strict emissions regulations.

   • Resource Shocks: A price shock or supply disruption.

   • Unusual Weather: A very warm winter reducing heating demand.

5. How could you use this Excel-based method to compare the resource consumption of two different countries?
   I would acquire the same time-series data for a second country, plot both lines on the same chart in Excel, and use the "Add Trendline" feature for each. This would allow for a direct visual and quantitative comparison of their consumption trends, growth rates (slopes), and volatility over the same time period.

Part B 

Aim

To analyze historical trends and patterns in natural resource consumption (e.g., energy, water, minerals) for a selected country using publicly available time-series data, and to interpret these trends in the context of economic, environmental, and policy factors.

---

1. Introduction

Natural resource consumption is a critical indicator of economic development and environmental impact. Time-series analysis allows us to examine trends, seasonal patterns, and structural changes in resource use over time. This practical will guide you through the process of acquiring, processing, and analyzing time-series data for a country of your choice, using statistical tools to draw insights into sustainable resource management

.

---

2. Learning Objectives

By the end of this practical, students will be able to:

• Identify and source publicly available time-series data on natural resource consumption.

• Preprocess and visualize time-series data to identify trends and patterns.

• Apply appropriate statistical models (e.g., ARIMA, exponential smoothing) to analyze and forecast resource consumption.

• Interpret results in the context of economic and environmental policies.

• Discuss limitations and ethical considerations of using public data for resource analysis.

---

3. Materials Required

• Computer with internet access.

• Statistical software (e.g., R, Python with pandas/statsmodels, or Excel).

• Public data sources (e.g., World Bank, UN Energy Statistics, BP Statistical Review, national agencies).

• Spreadsheet software for initial data exploration.

---

4. Procedure

Phase 1: Selection of Country and Resource

1. Choose a Country and Resource:

   • Select a country (e.g., Nepal, U.S., China) and a natural resource (e.g., natural gas, electricity, water)

2. • ).

Phase 2: Data Acquisition

1. Identify Data Sources:

   • Use platforms like:

     • World Bank Open Data (e.g., energy use, water withdrawals).

     • U.S. Energy Information Administration (EIA) (e.g., fossil fuel consumption).

     • BP Statistical Review of World Energy.

     • National agencies (e.g., China National Bureau of Statistics).

   • Ensure data is time-series (e.g., annual, monthly) and covers at least 20 years for robust analysis.

2. Data Collection:

   • Download data for the selected resource (e.g., natural gas consumption in billion cubic feet

2. • .

Phase 3: Data Preprocessing

1. Data Cleaning:

   • Handle missing values (e.g., interpolation or exclusion).

   • Adjust for units and inconsistencies.

2. Visualization:

   • Plot the raw data to identify trends, seasonality, and outliers.

   • Use graphs like line plots, seasonal decomposition plots, or autocorrelation plots

2. • .

Phase 4: Time-Series Analysis

1. Descriptive Analysis:

   • Calculate summary statistics (mean, variance, growth rates).

   • Discuss visible trends (e.g., increasing consumption due to industrialization

3. • .

Phase 5: Interpretation and Reporting

1. Contextualize Results:

   • Relate trends to historical events (e.g., policy changes, economic crises).

   • Discuss implications for sustainability (e.g., carbon emissions from fossil fuels

2. • ).

---

5. Observations and Data Analysis

Example: Natural Gas Consumption in the U.S. (2000–2024)

Data sourced from FRED

.

Year	Consumption (Bcf)	GDP (Trillion USD)	Population (Millions)
2000	1500	10.3	282
2010	2200	14.9	309
2020	3000	21.4	331
2024	3200 (forecast)	23.5 (projected)	336 (projected)

Key Steps:

1. Visualization:

   • Plot showing upward trend with seasonal fluctuations (higher in winter for heating).

2. Modeling:

   • ARIMA(2,1,0) model fitted to differenced data

2. • .

   • Forecast: Consumption stabilizes near 3200 Bcf by 2025.

3. Interpretation:

   • Growth driven by industrial demand and shift from coal.

   • Seasonal patterns align with heating demand.

---

6. Discussion

• Economic Drivers: Correlation between GDP growth and resource consumption (e.g., Nepal's energy-GDP nexus

• • .

---

7. Conclusion

Time-series analysis reveals complex dynamics in resource consumption, influenced by economic, technological, and policy factors. Public data, while accessible, requires careful handling to ensure robust insights. This practical underscores the importance of evidence-based analysis for sustainable resource management.

---

8. Viva Voce Questions

1. Why is time-series analysis particularly useful for studying natural resource consumption?

• It captures temporal patterns (trends, cycles) and allows forecasting, essential for policy planning 

       2. What are the key challenges in using public data for this analysis?

• Data gaps, inconsistent reporting, and aggregation biases may limit accuracy   

     3. How might climate change policies influence natural gas consumption trends in the future?

• Policies promoting renewables could reduce growth rates, as seen in forecasts   

     4. What alternative models could be used if the data shows nonlinear trends?

• Machine learning models (e.g., LSTM) or time-varying parameter models   

   5. How would you validate the forecast accuracy of your model?

• Use metrics like MAPE and compare against holdout data 

 

9. Additional Resources

• Data Repositories:

  • UCI Machine Learning Repository (time-series datasets).

  • NASA POWER (energy and climate data).

• Software Guides:

  • R: forecast package for ARIMA.

  • Python: statsmodels for time-series analysis.

---

This practical combines technical skills with critical thinking, preparing students to address real-world sustainability challenges. By following these steps, students will gain hands-on experience in data-driven environmental analysis.