DISCUSSION

RESEARCH QUESTIONS

1. How has the war in Yemen impacted night light cover?

2. How has the extent of cultivation changed in Western Yemen during the Civil War?

3. How has the health of Western Yemen's agriculture changed during the Civil War?

CONCLUSION

Analyzing night light cover is an effective method to study changes in human activity and development (or lack thereof). In this study, we were able to find night light imagery of Yemen before and after the 2015 bombings. Our night light detection map yielded the results we anticipated given considerable destruction of 2015. These results therefore illustrate the devastating impacts the war in Yemen has had on its civilians.

Since 2015, thousands of Yemenis have been killed and displaced. There have been many targeting bombings of churches, schools, hospitals, etc in 2015. For example, the El-Rahaba Airport and Al-Dulaimi Airbase located in Sana’a experienced a targeted bombing in to weaken the Houthis’ air power. This and the several other related bombings has greatly influenced human activity since 2012, as noted in the night light change detection map. Civilians continue to migrate to other parts of Yemen where resources might be less scarce. Migration patterns include from Sana'a to Amran and Hajjah to Al Hudaydah. The neighboring nations of Djibouti, Somalia, Saudi Arabia, Sudan, Ethiopia and Oman also continue to take refugees.

Another likely cause of the considerable night light reduction are the disease outbreaks of cholera and dengue that began in 2016. There is a significant lack of sanitation facilities and only 45% of health facilities are functioning, leaving 14.8 million Yemenis in need humanitarian aid. This reflects an increase in civilians lacking basic health and sanitation needs by 76% since late 2014. Mothers and children are left most vulnerable to these shortages. Areas identified by the UN as most in need of humanitarian aid (as of 2017) in Yemen include Al Hudayah, Sana'a, Taiz and Al Jawf. Most of these areas also experienced a considerable decline in night light cover.

Yemen is also highly dependent on food imports, and it is quite possible that the majority of the agricultural region identified in this study is dedicated to cash crop production. Therefore, agricultural productivity may not be tied to food availability very closely in Al-Hudaydah and that the famine is precipitated by external factors.

big_81d19fbf17ea2fa46b51e937f331daec325a

Challenges

1. Although we were only able to find night light data from 2012-2016 of Yemen, it would be useful to also have data of night lights today to see if night light cover has increased even more or not. This would allow for a better understanding of the current status of the country as the crisis is ongoing.

2. There is extremely limited data on Yemen’s agricultural system. The last nationwide agricultural survey was conducted in 1993, as a retrospective on the collapse of the cotton industry in the country at that time. Due to this lack of study, there has not ever been a land use classification produced for the country. This study did not include an investigation of the spectral signatures of the areas we classified as vegetation, and therefore in light of each of these three factors it is impossible to accurately state what crops might have been grown in the region of study.

3. The classification system used in this project was a simple band threshold classification, designating all NDVI above .13 as vegetation. Instead, a land use map could have been generated for this study, though that would take much more time. It is possible that, because of this, the study captured irrigated but fallow or weed-choked fields. The use of the difference map, however, should have balanced any issues from this to a large extent.

The crisis in Yemen continues with no clear indication as to when it will end and, therefore, continued monitoring of human activity and vegetation cover is essential to providing humanitarian aid to the country
Additionally, a more in-depth analysis of health systems and the spread of disease would be helpful to guide humanitarian groups in how to best address the disease outbreaks
Future areas of research on this topic would benefit greatly from a more comprehensive survey of the country’s agricultural system, starting with a land use map and classification of different crop types in the growing regions based on spectral signatures