Estimating of antibiotic resistance genes in the sediments of Erhai Lake, China: Based on multi-source remote sensing data

Antibiotic resistance genes (ARGs) in lakes have become significant emerging contaminants in recent years. The enrichment and dissemination of these genes within lake ecosystems pose potential threats to aquatic environments and human health. Traditional monitoring methods, such as high-throughput sequencing, are costintensive and unsuitable for large-scale or long-term monitoring. To address this challenge, this study proposes a method for estimating ARG abundance in inland lakes using multi-source remote sensing data. This study establishes regression models between Sentinel-2 MSI remote sensing reflectance and the concentrations of total phosphorus (TP) and total nitrogen (TN) in Erhai Lake. The performance of four regression models is compared to identify the most effective predictive approach. Additionally, the satellite altimetry data is integrated with the water balance equation to estimate the monthly runoff as well as the TN and TP pollutant loads discharged from tributaries into Erhai Lake from March 2018 to February 2019. A multivariate linear regression model is then developed to quantify the relationship between ARG abundance and pollutant loads. The result reveals that the linear relationship between total ARG abundance and the pollutant loads of TN and TP has an R2 of 0.759. TN input was positively correlated with ARG abundance, while TP input showed a negative correlation. And the spatial distribution of ARGs in Erhai Lake is successfully estimated. This study demonstrates the potential of multi-source remote sensing data as a powerful tool for estimating ARG abundance and offers a novel approach for large-scale and long-term monitoring of ARG spatial distribution in aquatic environments.