Back to success stories

Mapping land cover

The Regional Land Cover Monitoring System (RLCMS) addresses challenges in land management – including difficulties in accessing data, lack of transparency in data collection methodologies, inconsistencies in land cover classification, and limited financial and staff resources – by annually generating high-resolution land cover data for the HKH region.

70% Complete

High-resolution annual land cover data for the HKH region

Mapping land cover

The Regional Land Cover Monitoring System (RLCMS) addresses challenges in land management – including difficulties in accessing data, lack of transparency in data collection methodologies, inconsistencies in land cover classification, and limited financial and staff resources – by annually generating high-resolution land cover data for the HKH region. The system uses freely available remotesensing data and a cloud-based machine learning architecture to generate land cover maps through a harmonized and consistent regional classification system.

In 2019, we partnered with agencies in Afghanistan, Bangladesh, Myanmar, and Nepal to customize the RLCMS further as per national requirements, and conducted multiple trainings on the system’s development and use. In Nepal, the Forest Research and Training Centre (FRTC) has taken ownership, having allocated its own resources for field validation of the land cover data before final release. The system will be adopted for official reporting on forest cover and provide a basis for other forest-related applications such as national eco-region mapping. In Bangladesh, after a successful pilot in the Chittagong Hill Tracts the Bangladesh Forest Department (BFD) has rolled out the system for the entire country.

Early involvement of FRTC and BFD staff in the co-development of the system has helped build institutional capacities so that they can take the activity forward independently with limited technical backstopping from ICIMOD.

The RLCMS was developed through a joint collaboration among ICIMOD, Asian Disaster Preparedness Center (ADPC), United States Forest Services (USFS), and SilvaCarbon.

The system uses freely available remote-sensing data and a cloud-based machine learning architecture to generate land cover maps through a harmonized and consistent regional classification system.

butterfly

Chapter 2

Knowledge generation and use

Flood early warning saves lives

Ensuring that the right information reaches the right audience at the right time is crucial to reducing disaster impacts

Multi-scale integrated river basin management resource book co-creation reinforces its relevance

Demand-driven, context-specific workshop allows for module testing

Benchmark glaciers for monitoring in Afghanistan and Pakistan

Key steps towards more data generation, sharing and regional cooperation to understand and mitigate climate change impacts

Learning from a disaster event: Investigating the 2018 Panjshir flood in Afghanistan

In a case illustrative of effective inter-agency collaboration and resource sharing, the flash flood in Panjshir Valley, north-central Afghanistan, on ...

Engaging local-level policymakers in tailoring climate information

A rapidly changing climate and frequent extreme weather events are resulting in disturbances in the largely ...

Climate Services Initiative takes shape

Expanding knowledge on and access to climate and weather services

Municipal waste management policies underpin urban climate change adaptation

Our International Development Research Centre research grant-funded, transdisciplinary, multiinstitution research project – entitled ‘Cities and Climate ...

Science-based regional collaboration through the Upper Indus network 

Members are presently working on basin level issues focusing on climate change and resilience