Comparative Analysis of Drought in Africa and its Effects Since the 21st Century

Team Model-makers of the Hamoye Internships Fall 2023 Cohort

PROBLEM STATEMENT

The challenge we face is the imperative to gain a thorough understanding of the changing patterns of drought and their multifaceted impacts. This understanding is crucial for developing precise and effective mitigation strategies.

Data Source and Specification

For this analysis, we utilized a comprehensive dataset that includes information on drought occurrences in African countries since the year 2000. The dataset was taken from the EM-DAT page — https://public.emdat.be/graphql/files/public_emdat_custom_request_2023–11–08_c8191d8f-e11d-4a4e-86fe-6e1e591bea81.xlsx

The dataset comprised 168 rows and 46 columns, encompassing diverse parameters such as duration, intensity, and geographic location of drought events. Additionally, it included socio-economic and environmental data to evaluate the consequences of these events.

AIM AND OBJECTIVES

The aim of this study is to comprehensively analyze the distribution, severity and impacts of drought in African countries since the beginning of the 21st century. By comparing and contrasting these aspects across regions and time, the research seeks to provide valuable insights into the evolving nature of drought in Africa.

SOLUTION PROCESS / METHODOLOGY

The following steps were undertaken in the execution of the project:

• Data Inspection and Cleaning
• Data Analysis
• Data Visualization

Data Inspection and Cleaning:

Upon meticulous examination of the provided dataset, it became apparent that the data lacked sufficient substance for model training. Consequently, a decision was made to conduct a thorough analysis of the dataset. The data inspection process involved each team member trimming the dataset until a satisfactory and cleansed dataset was achieved. This process involved the following steps:

1. Handling Missing Data: The dataset contained missing data as well as null values which were inspected using the appropriate python code

2. Null columns, along with those containing a substantial number of null values, were dropped, as illustrated below.

3. Numeric cells with null values, such as the “Total Affected” column, were filled with zeros to prevent errors in calculations and enhance the accuracy of our analysis. Additionally, certain columns were filled with “Not Specified” for the same reason mentioned earlier.

4. The datatypes of the columns to be used were changed due to the kind of values they contained e.g. Start Year was changed from datatype “object” to “category” and so on.

5. Finally, the dataset “JME_Regional-Classifications.xlsx” was imported from the internet and relevant columns from the dataset were merged to our own dataset as shown:

Data Analysis (Exploratory Data Analysis)

The dataset contains information on the occurrence of drought in Africa. This analysis aims to answer the following:

Who were the worst hit? what is their distribution per: country, region and income group?

After completing the data cleaning process, we proceeded to conduct exploratory data analysis. This involved performing statistical computations and creating visualizations to address the problem statements outlined earlier in this documentation and to answer the research questions provided above. The exploratory data analysis consisted of two parts: Uni-variate and Multi-variate analysis. Both types of analyses were accompanied by various visualizations to provide deeper insights into the relationships between data items in the dataset.

Some are shown below:

Bar Plot of African countries vs the number of people affected

From the above analysis, the following countries were the most hit in terms of the number of persons affected by drought:

• Ethiopia
• Kenya
• Somalia
• Zimbabwe
• Nigeria.

Remarkably, the top three out of the five most affected countries still belonged to Eastern Africa. The question arises: Why is Eastern Africa predominantly affected by drought? Could it be attributed to geographical factors, or is it a consequence of inadequate government emergency responses? This query necessitates further investigation as our current dataset cannot provide a conclusive answer.

Additionally, various data visuals were generated, all of which depicted relationships between different data items in the dataset.

Analyzing the graph above and considering the unspecified values in the dataset, it is apparent that the most prevalent impact of drought is Food Shortage, closely followed by Famine. These conditions pose significant challenges for citizens, highlighting the urgency for government intervention to prevent migration from areas susceptible to this disaster.

The chart illustrates the frequency of drought occurrences between the years 2000 and 2022, revealing Somalia and Mozambique as the top-ranking countries, each experiencing drought 20 times. This surpasses the frequency in other African countries.

Bivariate Analysis

We integrated data from the “JME_Regional-Classifications.xlsx” dataset. This allowed us to classify drought distributions based on sub-regions defined by the United Nations. A pie chart was generated to visually represent this classification.

The pie chart clearly shows eastern Africa as the sub-region with the highest number of people affected by drought, the visual above shows that 54.9% of persons affected are from eastern Africa. This means that one in every two persons affected by drought in Africa is from eastern Africa.

The map above gives a distribution of drought occurrence in African countries according to severity level.

A bar chart was generated to illustrate the percentage of the total number affected by drought, categorized by income groups — low-income and high-income countries. The chart revealed that low-income countries experienced more significant impacts, exceeding those in middle-income countries by more than twice. This strongly indicates that a country’s income level plays a crucial role in the extent of its population affected by drought.

Furthermore, a plot of the Consumer Price Index (CPI) for these countries over the years demonstrated a consistent increase. This economic effect is typical in the aftermath of such disasters. Additional consequences, such as food shortage, crop failure, and famine, were also observed.

OBSERVATIONS / RECOMMENDATIONS

1. Changing Trends in Drought Occurrence: Our analysis uncovered dynamic trends in drought occurrences, showcasing both increases and decreases in different regions. Climate change indicators, notably rising temperatures and altered rainfall patterns, emerged as significant factors contributing to these shifts.
2. Regional Disparities: Drought severity varies widely across African regions. Countries in the sub-Saharan Africa have been particularly susceptible to severe and prolonged drought especially countries in east Africa, while other regions like Northern Africa have experienced milder events. This geographic variation necessitates region-specific drought management strategies.
3. Socio-economic Impact: Drought in the 21st century has resulted in significant socio-economic implications for many African nations. Negative impacts on economies and livelihoods are evident, including reduced agricultural productivity, loss of livestock, and heightened food insecurity.
4. Climate Change Link: Our analysis indicates a clear connection between climate change and heightened vulnerability to drought. The impact of rising temperatures and altered rainfall patterns exacerbates the likelihood and severity of drought, emphasizing the imperative for implementing climate adaptation measures.
5. Mitigation Strategies: African nations have undertaken various drought mitigation and adaptation strategies, including early warning systems, improved water resource management, and community-based resilience programs. Some of these measures have shown promise in reducing the impacts of drought.
6. Food Security Concerns: Drought has significantly impacted food security in African countries. The analysis revealed fluctuations in crop yields and a growing dependence on food aid during drought periods. This issue underscores the urgency of bolstering food security measures.

CONCLUSION

The comparative analysis of drought in Africa since the 21st century highlights the multifaceted nature of this environmental challenge. As climate change continues to exert its influence, the need for effective drought management strategies becomes more critical. The findings suggest that while the effects of drought vary across the continent, common threads include the socio-economic vulnerabilities and the importance of proactive measures.

This analysis serves as a valuable resource for policy-makers, researchers, and organizations aiming to address the consequences of drought in Africa. It underscores the necessity of region-specific interventions, climate resilience, and collaborative efforts to mitigate the impacts of drought in the 21st century and beyond. As we move forward, a holistic and data-informed approach is essential to build a more resilient and adaptive Africa in the face of a changing climate.