Although the article is very well written and base on a large data set, I have major concerns to be published in its present form. My main objection relates to the core conclusion of the paper, i.e. that a slight mean air temperature decrease along a period of 15 years of only 0.18 oC contributed to the improvement of biotic quality of river waters as a result of the enhancement of river water temperature. This conclusion led the authors to the further conclusion that “benthic invertebrate communities are able to recover from increasing temperatures” and to the assumption that “increases in water temperature will lead to opposite effects and therefore cause declining assessment results”. However, in my opinion the initial conclusion, i.e. that macroinvertebrate communities have recovered as a result of an air temperature decrease of 0.18 oC is not supported by the data for the following reasons:

It is well known that air and water temperature are strongly correlated. However, when considering rivers, water temperature is also affected by surface/groundwater interactions. Van Vliet et al. (2011), who studied global relationships between air and river water temperature, found that air temperature increases of +2, +4, and +6 oC, result in river water temperatures increases of +1.3, +2.6, and +3.8 oC, respectively. These findings are explained by the fact that river water temperature rise is probably buffered by the contribution of cooler ground water flow to river flow. It is thus expected that an air temperature difference of only 0.18 oC will practically not affect river water temperature or that river water temperature will rise so slightly that any respective changes in biogeochemical processes will not have measurable effects on biota assemblages.

It is well known that macroinvertebrate metrics are sensitive to different types of human pressures, including organic pollution, habitat modification etc. The mitigation of other stressors within the 1st and 2nd River Basin Management Plans through e.g. the implementation of better agricultural practices, diminishing of point pollution sources, or an improvement of hydromorphological conditions, or even an increased dilution effect through a respective increase in rainfall throughout the time of investigations are likely to be far more possible factors for the observed improvement of macroinvertebrate metrics.

Moreover, the authors applied a multitude of macroinvertebrate metrics that are not particular sensitive to temperature changes, and as they state “it is currently difficult to distinguish between a, for example, pollution signal and a temperature signal” (Lines 211-212). On the other hand, the CTI metric that they applied and as they state is temperature sensitive is not possible to be used alone to provide a scientific basis for their central statement.

Finally, the authors state “Air temperature was used as a proxy for water temperature due to the lack of water temperature measurements with adequate and comparably high temporal and spatial resolution…”. (Lines 406-408). When applying macroinvertebrate sampling, water temperature is routinely measured. How it is possible macroinvertebrate data and related metrics to be adequate (with high temporal and spatial resolution) for analyzing trends and, at the same time, water temperature data to be inadequate for statistical analysis?

I thus suggest to the authors to thoroughly revise their article by obtaining additional data and proceed to a more comprehensive analysis of the causes of the observed biotic-metric improvement, according to the following:

• Use water temperature data. • Obtain information regarding land-use changes and application of PoMs for the improvement of ecological quality for all river basins examined. • Consider and analyze long-term pollution data and trends. An improvement of the chemical-physicochemical quality or of the chemical status is expected to exert a much stronger influence on the improvement of biotic metrics than a slight water temperature decrease, if there is any. • Consider long-term precipitation data and analyze if there is a statistically accepted difference along the 15 years of investigations. If yes, a dilution effect could partly explain water quality improvements. In fact, the improvement of BMWP over time may reflect the aforementioned environmental changes. • Change Results and Conclusions within the Abstract section because they provide a wrong impression. • Make clear form the beginning of the paper that by “temperature” you mean “air temperature”. Otherwise the reader will gain a wrong impression. In fact, you have to go to the end of the paper (Methods) to get the information that all about temperature concerns air temperature and not water temperature. • Delete any statement related to the conclusion that temperature changes are the cause of respective biotic changes, e.g. “we demonstrate that temperature changes markedly affect community composition, as reflected by changes in various metrics” (Lines 186-187) and “Here, we showed significant changes in 10 out of 13 ‘composition and abundance metrics’, highlighting the direct connection between changes in the abundance of certain taxa and temperature” (Lines 213-215). It is not supported by the results.