1 Kunj glacier and glacial lake in Panjshir, Kabul basin, Afghanistan, in August 2018 (Photo: Mahboobullah Bariz/NWARA) The National Water Affairs Regulation Authority (NWARA) and ICIMOD worked together to develop information on glaciers and glacial lakes in Afghanistan through ICIMOD’s SERVIR Hindu Kush Himalaya (SERVIR-HKH) Initiative. Afghanistan’s glacier dataset (1990–2015) was released in 2018. The General Directorate of Water Resources of NWARA and ICIMOD then prepared datasets on glacial lakes in Afghanistan (1990–2015). This is a first-of-its-kind dataset on the distribution of glacial lakes in Afghanistan and observed changes since the 1990s that covers the whole of Afghanistan using consistent data sources and methods. ICIMOD developed the Glacial Lakes in Afghanistan application that provides an interactive visualization of the database on glacial lakes online. The interactive maps for glacial lakes were prepared for 1990, 2000, 2010, and 2015 using a uniform data set and methodology, which provide a scientific basis for understanding the changes taking place in the glacial environment in Afghanistan. LAUNCH APPLICATION DOWNLOAD INFORMATION SHEET 2 Distribution of glaciers and glacial lakes in Afghanistan Facts and figures Afghanistan has 1,942 glacial lakes that cover an area of 8 km2. The glacial lakes mostly range from 0.003 km2 to less than 1 km2 in Only three lakes are larger than 1 km2. The largest lake (14.63 km2) is located in the Upper Panj sub-basin. The glacial lakes are found from 2,900 masl to 5,400 masl; 91% of the lakes are located between 4,000 and 5,000 The Panj Amu basin contains the highest number (64%) and area coverage (74%) of glacial lakes in During a period of 25 years (1990–2015), the total number and area of glacial lakes in Afghanistan increased by 8% and 10%, In recent periods, the formation of new glacial lakes and the expansion of existing ones is higher than the disappearance and recession of glacial Methodology Landsat images were used to map glacial lakes from 1990, 2000, 2010, and 2015. The methodology used to map glaciers in Afghanistan, i.e. semiautomatic object-based image classification, was also used to map glacial lakes for consistency. Normalized difference water index (NDWI) [NDWI = (NIR-BLUE)/(NIR+BLUE)] was used as a primary index to categorize glacial lakes identified in the satellite images. Various filters – hue, normalized difference vegetation index (NDVI), and land and water mask (LWM) – were used to improve classification of glacial lakes misclassified as shadow areas resulting from spectral reflection, frozen lakes, fresh snow, and other land surfaces such as ice cliffs and walls of supraglacial lakes (Figure 2). The glacial lake polygons derived from automatic processes were further checked and refined manually by overlaying on Landsat imagery and cross-checking against high-resolution satellite images available in Google Earth. Finally, the glacial lakes were classified into eight different types based on dam characteristics and morphological forms described in Maharjan et al. (2018). Other parameters were generated using the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) and projecting the data in the World Geodetic System 1984 Universal Transverse Mercator projection. The glacial lake data for other periods (1990, 2000, and 2010) were prepared by manually correcting the 2015 lake data by draping over respective Landsat images. 3 Methodology of glacial lake mapping Status of glacial lakes in Afghanistan (2015) A total of 1,942 glacial lakes covering 88.8 km2 were mapped from the Landsat images of 2015. Glacial lakes greater than or equal to 0.003 km2 in size were considered for this inventory. These lakes are mostly concentrated in two river basins – Panj Amu and Kabul. Panj Amu basin has the highest number (64%) and area coverage (74% of total area) of the glacial lakes in Afghanistan. 4 Number and area percentage of glacial lakes in each sub-basin of Afghanistan Overall, more than 85% of the lakes are found within 5 km from glaciers in the country. Lake area coverage is also high within this distance. Almost 56% of the lakes are fed by glaciers. Lakes closer to glaciers (<5 km distance) are usually fed by glacial melt, whereas lakes that lie far ahead of glaciers may or may not be fed by glacial melt but instead are formed in paleo-glaciation landforms. Lakes closer to glaciers are mostly moraine dammed. The number and area of glacial lakes shows an increasing trend over the 25-year period (1990–2015), with a higher rate of increase during the second decade (2000-2010) and higher still in the last quinquennial (2010-2015). More new glacial lakes have formed than have disappeared, and more existing lakes have expanded than have receded. The rate of unchanged lakes has decreased in recent years, which indicates an increasing trend in the formation and expansion of lakes. This calls for systematic monitoring of these lakes. A detailed investigation should also be carried out to identify potentially dangerous glacial lakes to reduce the risk from glacial lake outburst floods (GLOFs). About SERVIR SERVIR connects space to village by helping developing countries use satellite data to address challenges in food security, water resources, weather and climate, land use, and natural disasters. A partnership of NASA, USAID, and leading technical organizations, SERVIR develops innovative solutions to improve livelihoods and foster self-reliance in Asia, Africa, and the Americas. ICIMOD implements the SERVIR-HKH Initiative – one of five regional hubs of the SERVIR network – in its regional member countries, prioritizing activities in Afghanistan, Bangladesh, Myanmar, Nepal and Pakistan. SERVIR-HKH has specific priorities in Afghanistan related to the thematic areas of agriculture and food security; land cover, land use and ecosystems; and water resources and hydro-climatic disasters supported by USAID-Afghanistan. The initiative aims to support the Government of Afghanistan’s efforts in capacity development, data accessibility and availability, and enhancing provision of user-tailored data and tools for decision-making in a sustainable manner.