SUMMARY: The crop growth model HERMES was used to model crop rotation cycles at 12 experimental sites in the Czech Republic. A wide range of crops (spring and winter barley, winter wheat, maize, potatoes, sugar beet, winter rape, oats, alfalfa and grass), cultivated between 1981 and 2009 under various soil and climatic conditions, were included. The model was able to estimate the yields of field crop rotations at a reasonable level, with an index of agreement (IA) ranging from 0·82 to 0·96 for the calibration database (the median coefficient of determination (R²) was 0·71), while IA for verification varied from 0·62 to 0·93 (median R² was 0·78). Grass yields were also estimated at a reasonable level of accuracy. The estimates were less accurate for the above-ground biomass at harvest (the medians for IA were 0·76 and 0·72 for calibration and verification, respectively, and analogous medians of R² were 0·50 and 0·49). The soil mineral nitrogen (N) content under the field crops was simulated with good precision, with the IA ranging from 0·49 to 0·74 for calibration and from 0·43 to 0·68 for verification. Generally, the soil mineral N was underestimated, and more accurate results were achieved at locations with intensive fertilization. Simulated yields, soil N, water and organic carbon (C) contents were compared with long-term field measurements at Němčice, located within the fertile Moravian lowland. At this station, all of the observed parameters were reproduced with a reasonable level of accuracy. In the case of the organic C content, HERMES reproduced a decrease ranging from c. 85 to 77 tonnes (t)/ha (for the 0–0·3 m soil layer) between the years 1980 and 2007. In spite of its relatively simple approach and restricted input data, HERMES was proven to be robust across various conditions, which is a precondition for its future use for both theoretical and practical purposes.

Freshwater fish production is significantly correlated with water temperature, which is expected to increase under climate change. This study evaluated changes in water temperature and their impact on productive ponds at fisheries in the Czech Republic. A model was developed to calculate surface water temperature based on the five-day mean of the air temperature and was then tested in several ponds in three major Czech fish production areas. The output of the surface water temperature model was compared with independently measured data (r = 0.79–0.96), and the verified model was then applied to predict climate change conditions. The results were evaluated with regard to the thresholds characterizing the water temperature requirements of fish species and indicated that the limitation of Czech fish farming results from (i) an increased number of continuous periods during which given fish species are threatened by high water temperatures and (ii) the extension of continuous periods with stressful water temperatures. For Czech fisheries, the model suggests a sharp increase in unprecedented temperature regimes, which will pose critical challenges to traditional forms of common carp farming within several decades. Although reducing the level of eutrophication and loading them with organic substances might alleviate expected threads, farming current fish species in deeper and colder ponds at higher elevations might be inevitable.