Decoupling the heterogeneity of sediment microbial communities along the urbanization gradients: A Bayesian-based approach

Abstract

Comprehending the response of microbial communities in rivers along urbanization gradients to hydrologic characteristics and pollution sources is critical for effective watershed management. However, the effects of complex factors on riverine microbial communities remain poorly understood. Thus, we established a bacteria-based index of biotic integrity (Ba-IBI) to evaluate the microbial community heterogeneity of rivers along an urbanization gradient. To examine the response of Ba-IBI to multiple stressors, we employed a Bayesian network based on structural equation modeling (SEM-BN) and revealed the key control factors influencing Ba-IBI at different levels of urbanization. Our findings highlight that waterborne nutrients have the most significant direct impact on Ba-IBI (r = -0.563), with a particular emphasis on ammonia nitrogen, which emerged as the primary driver of microbial community heterogeneity in the Liuyang River basin. In addition, our study confirmed the substantial adverse effects of urbanization on river ecology, as urban land use had the greatest indirect effect on Ba-IBI (r = -0.460). Specifically, the discharge load from wastewater treatment plants (WWTP) was found to significantly negatively affect the Ba-IBI of the entire watershed. In the low urbanized watersheds, rice cultivation (RC) and concentrated animal feeding operations (CAFO) are key control factors, and an increase in their emissions can lead to a sharp decrease in Ba-IBI. In moderately urbanized watersheds, the Ba-IBI tended to decrease as the level of RC emissions increased, while in those with moderate RC emissions, an increase in point source emissions mitigated the negative impact of RC on Ba-IBI. In highly urbanized watersheds, Ba-IBI was not sensitive to changes in stressors. Overall, our study presents a novel approach by integrating Ba-IBI with multi-scenario analysis tools to assess the effects of multiple stressors on microbial communities in river sediments, providing valuable insights for more refined environmental decision-making.

Keywords: Anthropogenic contamination; Ba-IBI; Bayesian networks; Microbial community; Urbanization.