Remote sensing to quantify ecosystem services for monitoring and evaluating ecological restoration interventions
Trini del Rio is a PhD student in the department of Natural Resources. (Co)Promotors are prof.dr.ir. L.L.J.M. Willemen, prof. A.D. Nelson and dr.ir. A. Vrieling from the Faculty Geo-information and Earth Observation (ITC).
Ecological restoration is understood as 'the process of assisting in the recovery of an ecosystem that has been degraded, damaged, or destroyed'. In the last two decades, there has been growing recognition of the importance of ecological restoration in achieving sustainable development goals and addressing climate change, leading to a substantial increase in the global area designated for ecological restoration. The current UN Decade on Ecosystem Restoration (2021 – 2030) is a global initiative to accelerate action towards restoring degraded and destroyed ecosystems on a large scale. Many countries and organizations have committed to large-scale restoration initiatives.
Restoring degraded ecosystems is a challenging undertaking that demands substantial funds, effort, and expertise. Nonetheless, the process of restoration can generate both direct and indirect economic benefits, such as creating employment opportunities, enhancing land productivity, and reducing costs associated with natural disasters. Monitoring ecosystem services, 'nature's benefits to people,' for evaluating these restoration efforts is crucial for quantifying whether and when the intervention generates the desired environmental and social benefits. It is important to know where the intervention works best, to learn from past experiences, and to adjust and refine restoration strategies over time.
To advance our understanding of the effects of restoration initiatives, we need spatially explicit, frequent, and reliable data on land resources. However, monitoring resources over large areas can be costly and labor intensive. Fortunately, the use of remote sensing can help to bridge the data gap that often limits the evaluation of large-scale restoration interventions, and can reduce labor-intensive monitoring efforts in the field. In the four core chapters of this thesis (Chapters 2 to 5), a methodological approach was developed and applied that to evaluate the effect of restoration interventions on the spatial-temporal dynamics of ecosystem services using optical remote sensing. First, models were built to link remote sensing-derived indices to ecosystem services indicators on the ground. Second, the intra-annual variability of ecosystem services supply related to their demand was studied. Third, combining the BACI (Before-After, Control-Impact) analysis with remote sensing, pixel-level variability in the supply of ecosystem services was mapped and linked to the effect of ecological restoration efforts. The spatial distribution of restoration outcomes was also assessed in relation to terrain characteristics. Finally, the impact of remote sensing-related choices on the outcomes of the evaluation of restoration interventions was explored.
The new findings resulting from this thesis are as follows: The developed remote sensing-based models allow for the estimation of ecosystem service supply for large-scale monitoring. The models can assess ecosystem service supply more accurately when their indicators mainly depend on green vegetation, such as erosion prevention and provision of forage. The approach allows to assess intra-annual changes in the supply of ecosystem services at the pixel level. The relation between intra-annual supply and demand shows when ecosystem services are most needed during the year. The BACI method applied with remote sensing data can indicate changes in ecosystem service supply due to restoration interventions at the pixel level, allowing for the distinction of restoration effects from other factors. Additionally, the approach can help identify some of the factors affecting spatial differences in restoration outcomes. This thesis found that terrain, aspect, and soil parent material could have an impact on restoration success. Choices of remote sensing data, image timing, and methods need to be considered carefully, as they can affect conclusions about intervention impacts. From the explored remote sensing-related choices, this thesis found that the reference period and intra-annual image selection have a strong effect on restoration evaluation outcomes.
The results of this study can contribute to guiding the design of more efficient and timely monitoring and evaluation for large ecological restoration interventions. In the last chapter of this thesis (Chapter 6), the detailed contributions of this thesis are discussed in relation to the state-of-the-art of remote sensing developments. It also highlights the challenges and opportunities of the role of remote sensing for monitoring ecological restoration interventions and discusses prospects for implementing remote sensing-based monitoring and evaluation of such interventions. Regardless of the substantial progress in technology, research, data availability and opportunities for learning skills, there is still a gap between the potential of current remote sensing technologies for monitoring and their use by stakeholders in restoration initiatives. Remote sensing approaches for evaluation, such as the BACI analysis presented in this study, can substantially lower labor costs. However, ground measurements throughout the whole restoration project are still important for validation and calibration and remain one of the main constraints for effective support by remote sensing for monitoring the impact of restoration interventions.
