The Use of Remote Sensing Analysis for Evaluating the Impact of Development Projects in the Yellow Sea Large Marine Ecosystem

Abstract

Remote sensing has long been valued as a data source for monitoring environmental indicators and detecting trends in ecosystem stress from anthropogenic causes such as deforestation, river dams and air and water pollution. More recently, remote sensing analyses have been applied to evaluate the impacts of environmental projects and programs on reducing environmental stresses. Such evaluation has focused primarily on the change in above-surface vegetation such as forests. This study uses remote sensing ocean color products to evaluate the impact on reducing marine pollution of the Global Environment Facility’s (GEF) portfolio of projects in the Yellow Sea Large Marine Ecosystem. Chlorophyll concentration was derived from satellite images over a time series from the 1990s, when GEF projects began, until the present. Results show a 50% increase in chlorophyll until 2011 followed by a 34% decrease until 2019, showing a potential delayed effect of pollution control efforts. The rich time series data is a major advantage to using geospatial analysis for evaluating the impacts of environmental interventions on marine pollution. However, one drawback to the method is that it provides insights into correlations but cannot attribute the results to any particular cause, such as GEF interventions.

Evaluating site selection at design in food systems interventions: A formative geospatial approach

“…Because they are performed before program completion, formative evaluations lack quantitative evidence on results, especially when conducted early in implementation before monitoring data become available. This is especially true for evaluations of programs aiming to reduce environmental stress, since such results may only become apparent many years after program completion (GEF IEO, 2018a; Sidman et al, 2020). Quantitative evidence, especially if coming from datasets created by outside sources, reduces subjectivity, bias or lack of transparency if evidence sources need to be hidden for confidentiality, which can be inherent in qualitative methods (Ramona, 2011).…”

“…Within evaluation, geographic information systems (GIS) have primarily been used for data visualization, although GIS is also used widely for spatial analysis (Kates and Coryn, 2021). One common use of spatial analysis in environmental evaluation is utilizing remote sensing to measure change in environmental indicators before and after an intervention or between an intervention area and a control area, such as change in deforestation rates or vegetation indices (Kerle et al, 2019; Lech et al, 2018; Sidman et al, 2020). Others have used spatial modeling combined with household survey data to estimate an intervention’s impact on socioeconomic indicators (Runfola et al, 2020).…”