Risk based management of aquatic resources for ecosystem and public health requires water managers and health professionals to work together. Using an epidemiologic time-series modelling approach, we assess patterns of risk for alert-level cyanobacterial abundance with water temperature. We focus on six sites along the Murray–Darling Drainage Basin, using the longest continuous record of algal abundance in Australia. Alert-level cyanobacterial abundance showed a non-linear and lagged response to water temperature across all six sites, after controlling for relative water discharge. For three sites there was a positive relationship of high-water temperature with the risk of alert-level abundance. These three sites also showed a substantial lagged effect, with the risk remaining high at a lag of 1 month following high water temperatures. The higher than average risk of alert-level cyanobacterial abundance with extreme water temperature and the persistence of this effect for 1 month highlight the applicability of these models to understand non-linear and time-dependent relationships in complex systems which are managed for ecosystem and population health. The site-specific relationships provide guidance for local authorities to develop water quality-related environmental and public health responses to a variable climate.

基于风险的生态系统和公共卫生水生资源管理要求水管理人员和卫生专业人员共同努力。使用流行病学时间序列建模方法，我们评估了随水温升高的预警级蓝细菌数量的风险模式。我们利用澳大利亚最长的藻类持续记录，重点研究了默里-达令流域的六个地点。在控制了相对水排放之后，所有六个位置的警戒级蓝细菌丰度均显示出对水温的非线性和滞后响应。对于三个地点，高水温与警报水平丰度的风险呈正相关。这三个地点也显示出明显的滞后效应，在水温升高后的1个月内，风险仍然很高。极端水温下警报水平的蓝细菌丰度的风险高于平均水平，并且这种影响持续了1个月，突显了这些模型在理解针对生态系统和种群进行管理的复杂系统中非线性和时间相关关系的适用性健康。特定地点的关系为地方当局提供指导，以制定与水质有关的环境和公共卫生应对可变气候的措施。