Using Multisensor Earth Observation Data to Understand the Development of the Kourtéré Kori Alluvial Fan and Its Impact on the Hydrology of the Niger River

Abstract

The Niger River (NR) is threatened with silting up due to the evolution of alluvial fans (AFs) in the Niamey region. The most significant is the Kourtéré AF. This study aims to map the spatial extent of this AF and analyze its progression and impact on the NR using a machine learning algorithm such as support vector machine on Landsat and Sentinel-2 images from between 1985 and 2021. Sentinel-1 data were then used to generate a digital terrain model at 10 m resolution through the Interferometric Synthetic Aperture RADAR (InSAR) technique to understand the Kourtéré watershed hydrology. Results show that the AF area increased from 55.19 ha in 1985 to 303.86 ha in 2003 and 237.18 ha in 2021. The AF has invaded a section of the NR since 2000 and began to invade its main bed in 2016. More than 73.71 ha of NR were converted into AF between 1985 and 2016. The flow direction of the NR has been disturbed by changes in AF, making it easier to overflow at flows much lower than previously. The Kourtéré watershed is dominated by a dense network of gullies with a Sediment-Transport Index of 424.4, which facilitates significant sediment transport into the AF and NR and leads to its silting up.

尼亚美地区冲积扇的演变, 导致了尼日尔河(尤其是Kourtéré冲积扇)面临淤积的威胁。本研究旨在绘制Kourtéré冲积扇的空间范围, 并利用机器学习算法分析它的演变及其对尼日尔河的影响。例如, 利用支持向量机, 本文分析了1985年至2021年的Landsat和Sentinel-2影像。通过干涉合成孔径雷达(InSAR)技术, 用Sentinel-1数据生成10米分辨率的数字地形模型, 以了解Kourtéré流域的水文情况。结果表明, 冲积扇面积从1985年的55.19公顷增加到2003年的303.86公顷和2021年的237.18公顷。自2000年以来, 冲积扇侵入了尼日尔河的部分区域, 并于2016年开始侵入尼日尔河的主河床。1985年至2016年, 超过73.71公顷的尼日尔河被转化为冲积扇。受到尼日尔河变化的干扰, 尼日尔河的流向更易于在比以前低得多的流量下溢出。Kourtéré流域以密集的冲沟网络为主, 泥沙输送指数为424.4, 促进了向冲积扇和尼日尔河输送大量泥沙并导致淤积。

El Río Níger (NR) sufre la amenaza de sedimentación debido a la evolución de los abanicos aluviales (AFs) en la región de Niamey. El más importante de éstos es el Kourtéré AF. Este estudio apunta a cartografiar la extensión espacial de este AF y a analizar su progresión e impacto sobre el NR usando un algoritmo de aprendizaje automático como el support vector machine, sobre imágenes Landsat y Sentinel-2, entre 1985 y 2021. Los datos de Sentinel-1 se usaron posteriormente para generar un modelo digital del terreno a una resolución de 10 m por medio de la técnica del RADAR de Apertura Sintética Interferométrica (InSAR), para así entender la hidrología de la cuenca de Kourtéré. Los resultados muestran que el área del AF se incrementó de 55,19 ha en 1985 a 303,86 ha en 2003 y 237,18 ha en 2021. El AF ha invadido una sección del NR desde el 2000 y empezó a invadir su lecho principal en 2016. Más de 73,71 ha del NR fueron convertidas en AF entre 1985 y 2016. La dirección del flujo del NR ha sido afectada por los cambios del AF, facilitando aún más el desbordamiento con caudales más bajos a los anteriormente registrados. La cuenca del Kourtéré está dominada por una densa red de cárcavas, con un Índice de Sedimento-Transporte de 424,4, lo cual facilita un mayor transporte de sedimentos hacia el AF y el NR, y conduce a su colmataje.

Key Words:

• alluvial fan
• Kourtéré kori
• machine learning
• Niamey/Niger
• support vector machine

关键词：:

• 冲积扇
• Kourtéré Kori
• 机器学习
• 尼亚美/尼日尔
• 支持向量机。

Palabras clave::

• aprendizaje automático
• abanico aluvial
• Kourtéré kori
• Niamey/Níger
• support vector machine

Acknowledgment

Mamane Barkawi Mansour Badamassi is now with Drone Service, Niger.

Additional information

Notes on contributors

Mamane Barkawi Mansour Badamassi

MAMANE BARKAWI MANSOUR BADAMASSI is an independent earth observation researcher specializing in precision agriculture. He is the head of the 3D department (Geomatic, Drone, Lidar, BIM, Precision Agriculture) at XY CONCEPT in Rabat, Morocco. E-mail: [email protected]. His research is all about use of remote sensing, GIS, and artificial intelligence in several thematic areas to provide sustainable development solutions.

Konate Moussa

KONATE MOUSSA is a Professor in Structural Geology in the Faculty of Sciences and Technology, Abdou Moumouni University, Niamey, Niger. E-mail: [email protected]. His research focuses on two topics: Paleoproterozoic and pan-African basement areas of Niger: structural evolution and mineral resources, and geodynamic context of the geological and hydrogeological resources of the sedimentary basins of Niger.

Labaly Touré

LABALY TOURÉ is a Teacher-Researcher at Sine Saloum Elhadj Ibrahima Niass University, Senegal. E-mail: [email protected]. His research focuses on the themes of applied geomatics.

El-Aboudi Ahmed

EL-ABOUDI AHMED is a Professor of Biology and Applied Remote Sensing in the Department of Biology, Faculty of Science, Mohammed University, Rabat, Morocco. E-mail: [email protected]. His research concerns the study of ecosystems, biodiversity, and forests using remote sensing and GIS.

Gbetkom Paul Gerard

GBETKOM PAUL GERARD is a Postdoctoral Researcher at Midi-Pyrenees Observatory, Toulouse, France. E-mail: [email protected]. His research concerns the study of surface water resources, particularly on Lake Chad.

Hanadé Houmma Ismaguil

HANADÉ HOUMMA ISMAGUIL is a PhD Student in artificial intelligence in the Department of Geodesy and Topography, Geomatics, Science and Engineering, Hassan II Institute of Agronomy and Veterinary, Rabat, Morocco. E-mail: [email protected]. His research focuses on the use of artificial intelligence to monitor agricultural drought.

Kabore Gérard

KABORE GÉRARD is the Chief of Staff of the Executive Secretary at the Niger Basin Authority (NBA), Niamey, Niger. E-mail: [email protected]. Its mission is to contribute to the monitoring, management, and preservation of the Niger River.