The researchers say that paying attention to the errors could improve fusion remote monitoring of lakes.

Lakes can quickly go from healthy to potential environmental hazard when they become eutrophic, a condition in which abundant nutrients accelerate the growth of algae, which then crowd the surface and block light from reaching organisms below.

Without light, oxygen can’t be produced, and life in the water would begin to die. Fortunately, researchers can use remote sensing techniques to monitor eutrophication in inland lakes, but China-based researchers say these techniques could be adjusted to provide a more accurate assessment.

The team published their recommendations for evaluating the technology and improving it on September 3 in the Journal of Remote Sensing.

Current technology consists of remote sensing instruments capturing features on the Earth’s surface, known as spatial resolution, and capturing the same features multiple times, known as temporal resolution. The more detailed the image, and the more frequently the image is repeated, the higher the resolution. However, there is a trade-off between resolutions: the higher the spatial resolution, the lower the temporal resolution tends to be, and vice versa.

“The trade-off between spatial and temporal resolution of remote sensing instruments limits our ability to monitor the eutrophication status of inland lakes,” said co-corresponding author Linwei Yu, an associate professor at China University of Geosciences.

“Space-time fusion (STF) provides a cost-effective method to generate remote sensing data with both high spatial and temporal resolution by fusing multiple sensor information, and is widely used for detailed monitoring of the Earth’s surface dynamics.”

But the problem, the researchers say, is that processing and modeling errors in fusion monitoring can affect image quality, especially when photographing reflective surfaces such as lakes, which have a relatively low signal-to-noise ratio (SNR) – this ratio refers to the difference between relevant information and other details.

“This study preliminarily presents a comprehensive assessment to understand the potential and limitations of applying STF technology to monitor chlorophyll a (Chla) concentrations in inland eutrophic lakes,” said co-corresponding author Huanfeng Shen, a professor at Wuhan University, explaining that Chla is an indicator of eutrophication status.

“The results of this study will help provide guidelines for designing an STF framework for monitoring inland aquatic environments using remote sensing data.”

The researchers found that the STF method effectively captures the highly dynamic conditions of eutrophic inland lakes, but that those assessing the images need to pay “particular attention” to sources of error.

“Among the influencing factors, atmospheric correction and geometric errors have a significant impact on the fusion results,” Yu said. “We recommend a practical pipeline so that the fused images can be integrated with real observations to create a temporally dense Chla dataset.”

According to the researchers, the working pipeline detailed in the study provides a comprehensive understanding of the possibilities and uncertainties of using the STF method for aquatic applications.

“With this understanding, it will be possible to estimate temporally dense Chla concentrations in inland eutrophic lakes by fusing observations from multiple sensors,” Shen says. “In future studies, our goal is to integrate data from sensors with different resolutions to generate a Chla dataset with both high spatial and temporal resolution for large lakes.”