As climate change warms rivers, they lose dissolved oxygen from the water. This process, which is called deoxygenation , was already known to occur in large bodies of water such as oceans and lakes. A study that my colleagues and I have just published in Nature Climate Change,  shows that this also happens in rivers. We documented this change using a type of artificial intelligence called a deep learning model—specifically model of long-term short-term memory – to predict water temperature and oxygen levels. The data we fed into the model included past records of water temperature and oxygen concentrations in the rivers, along with past weather data and the characteristics of the surrounding land – such as whether it contained cities, farms or forests. However, the original data on water temperature and oxygen were measured sparsely and often in different periods and with different frequency. Therefore, before our study, it was difficult to compare across rivers and in different periods. ( Li Li (李黎) , Penn State)