Forests cover approximately 30% of the Earth’s land surface and have played an indispensable role in the human development and preserving natural resources. At the moment, more than 300 million people are directly dependent on these forests and their resources. Forests also provide habitats for a wide variety of species and offer several ecological necessities to natural and anthropological systems. In spite of this importance, unprecedented destruction of tropical forest cover has been witnessed over the past four decades. Annually, approximately 2.1x105 hectares of forests are lost, with serious negative consequences on the regulation of the world’s climate cycle, biodiversity and other environmental variables. To mitigate these consequences, the United Nations Framework Convention on Climate Change (UNFCCC) has requested the developing countries to adapt new policy in reducing emissions from deforestation and forest degradation (REDD+). Under this policy, countries have been mandated to engage local communities and indigenous groups as critical stakeholders in the design and implementation of a national forest monitoring system (NFMS) that supports measuring, reporting and verification (MRV) of actions and achievements of REDD+ activities.
Current schemes for tropical monitoring are based on remote sensing and field measurements which typically originate from national forest inventories. Remotely sensed imagery has been considered as the principal data source used to calculate forest area change across large areas, assess rates of deforestation and establish baselines for national forest area change databases. Advancements in medium and high resolution satellites, open data policies, time-series analysis methods and big data processing environments are considered valuable for deforestation monitoring at local to global scales. However, cloud cover, seasonality and the restricted spatial and temporal resolution of remote sensing observations limits their applicability in the tropics. Enhancing the interpretation of remote sensing analysis require substantial ground verification and validation. Accomplishing these tasks through national forest inventory data is expensive, time-consuming and difficult to implement across large spatial scales.
Next to remote sensing, community-based monitoring (CBM) has also demonstrated potential in the collection and interpretation of forest monitoring data. However effective implementation of community-based forest monitoring systems is currently lacking due to two reasons: 1) the role of communities in NFMS is unclear and 2) tools that can support local communities to explore opportunities and facilitate forest monitoring are still scarce. This thesis addresses these two issues by proposing technical solutions (computer and geo-information science) and assessing the capacities and needs of communities in developing countries with a REDD+ implementation and forest monitoring context.
The main goal of this thesis, therefore, is to develop an approach that combines emerging technologies and community-based observations for tropical forest monitoring. To accomplish the main goal, four specific research questions were formulated: 1) What are the potentials to link community-based efforts to national forest monitoring systems? 2) How can information and communication technologies (ICTs) support the automation of community data collection process for monitoring forest carbon stocks and change activities using modern handheld devices? 3) What is the accuracy and compatibility of community collected data compared to other data (e.g., optical remote sensing and expert field measurements) for quantifying forest carbon stocks and changes? and 4) What is a suitable design for an interactive remote sensing and community-based near real-time forest change monitoring system and how can such system be operationalized?
In Chapter 2, scientific literature and 28 readiness preparation proposals from the World Bank Forest Carbon Partnership Facility are reviewed to better define the role and technical conditions for CBM. Based on this review, a conceptual framework was developed under which CBM can contribute as a dedicated and independent stream of measuring and monitoring data to national level forest monitoring efforts. The following chapters are built upon this framework.
Chapter 3 describes a process of designing and implementing an integrated data collection system based on mobile devices that streamlines the community-based forest monitoring data collection, transmission and visualization process. The usability of the system is evaluated in the Tra Bui commune, Quang Nam province, Central Vietnam, where forest carbon and change activities were measured by different means such as local, regional and national experts and high resolution satellite imagery. The results indicate that the local communities were able to provide forest carbon measurements with accuracy comparable to that of expert measurements at lower costs. Furthermore, the results show that communities are more effective in detecting small scale forest degradation caused by subsistence fuelwood collection and selective logging than image analysis using SPOT imagery.
To support the findings of chapter 3, the data acquisition form (mostly activity data related to forest change) for mobile device was further improved in chapter 4. The system was tested by thirty local experts in the UNESCO Kafa Biosphere Reserve, Ethiopia. High resolution satellite imagery and professional measurements were combined to assess the accuracy and complementary use of local datasets in terms of spatial, temporal and thematic accuracy. Results indicate that the local communities were capable of describing processes of change associated with deforestation, forest degradation and reforestation, in terms of their spatial location, extent, timing and causes within ten administrative units. Furthermore, the results demonstrate that communities offer complementary information to remotely sensed data, particularly to signal forest degradation and mapping deforestation over small areas. Based on this complementarity, a framework is proposed for integrating local expert monitoring data with satellite-based monitoring data into a NFMS in support of REDD+ MRV and near real-time forest change monitoring.
Having identified the framework for integrated monitoring systems in chapter 4, chapter 5 describes an interactive web-based forest monitoring system using four levels of geographic information services: 1) the acquisition of continuous data streams from satellite and community-based monitoring using mobile devices, 2) near real-time forest disturbance detection based on satellite time-series, 3) presentation of forest disturbance data through a web-based application and social media and 4) interaction of the satellite-based disturbance alerts with the end-user communities to enhance the collection of ground data. The system was developed using open source technologies and has been implemented together with local experts in UNESCO Kafa Biosphere Reserve, Ethiopia. The results show that the system was able to provide easy access to information on forest change and considerably improve the collection and storage of ground observation by local experts. Social media lead to higher levels of user interaction and noticeably improved communication among stakeholders. Finally, an evaluation of the system confirmed its usability in Ethiopia.
Chapter 6 presents the final conclusions and provides recommendations for further research. The overall conclusion is that the emerging technologies, such as smartphones, Web-GIS and social media, incorporated with user friendly interface improve the interactive participation of local communities in forest monitoring and decrease errors in data collection. The results show that CBM can provide data on forest carbon stocks, forest area changes as well as data that help to understand local drivers of emissions. The thesis also shows, in theory and in practice, how local data can be used to link with medium and high resolution remote sensing satellite images for an operational near real-time forest monitoring system at a local scale. The methods presented in this thesis are applicable to a broader geographic scope. Hence, this thesis emphasizes that policies and incentives should be implemented to empower communities and to create institutional frameworks for community-based forest monitoring in the tropics.