Claudia Libiseller

Trends in nutrient concentrations in Latvian rivers and the response to the dramatic change in agriculture

In recent years, the use of fertilisers in the Baltic countries (Estonia, Latvia, and Lithuania) has decreased at an unprecedented rate. The import of mineral fertilisers and feed stuff became almost non-existent, and extensive slaughtering of livestock reduced the amount of manure. In Latvia, the purchase of mineral fertilisers decreased by a factor of 15 between 1987 and 1996 and the number of livestock decreased with a factor of almost 4 during the same time period. Such abrupt and comprehensive changes in land use have never before occurred in the history of modern European agriculture. Here, the impact that this dramatic reduction has had on concentrations of nutrients in Latvian rivers is examined. To discern temporal changes, statistical analyses were undertaken on time series of nutrient concentrations and relationships between concentrations and runoff at 12 sampling sites in ten Latvian rivers covering drainage areas from 334 to 64,000 km 2 . Considering the study period 1987‐ 1998, only four of the 12 sites showed statistically significant downward trends (one-sided test at the 5% level) in the dissolved inorganic nitrogen (DIN ¼ NO3-N þ NO2-N þ NH4-N) data. There are probably two main explanations for the weak DIN trends. Firstly, long water-transit times in the soilwater and groundwater may have caused substantial time lag between changes in input and output of nitrate in the studied catchments. Secondly, the loss of DIN might have been dominated by mineralisation of large pools of organic nitrogen that have accumulated over several years. These inferences are supported by (i) a hydrograph recession analysis and (ii) indications of DIN transformation processes, presumably denitrification, in smaller streams and channels, based on measurements in small agricultural catchments (1 ‐ 4 km 2 ) in Estonia and Latvia. Formal testing of trends in phosphorus data revealed that marked drops occurred in riverine concentrations at six sites in 1987‐ 1998. A joint analysis of concentration time series for all sampling sites for 1987‐ 1998 showed weak statistical significance for downward trends in NH4-N, NO3-N, and DIN ðp o 0:04Þ and substantial significance for PO4-P ðp , 0:01Þ: Thus, the extensive decrease in agricultural intensity that began in the early 1990s has led to only a slow and limited (especially regarding nitrogen) response in Latvian rivers. The difference noted between nitrogen and phosphorus also suggests that factors other than reduced fertiliser application influenced the inertia of the water quality response.

Model selection for local and regional meteorological normalisation of background concentrations of tropospheric ozone

Meteorological normalisation of time series of air quality data aims to extract anthropogenic signals by removing natural fluctuations in the collected data. We showed that the currently used procedures to select normalisation models can cause over-fitting to observed data and undesirable smoothing of anthropogenic signals. A simulation study revealed that the risk of such effects is particularly large when: (i) the observed data are serially correlated, (ii) the normalisation model is selected by leave-one-out cross-validation, and (iii) complex models, such as artificial neural networks, are fitted to data. When the size of the test sets used in the cross-validation was increased, and only moderately complex linear models were fitted to data, the over-fitting was less pronounced. An empirical study of the predictive ability of different normalisation models for tropospheric ozone in Finland confirmed the importance of using appropriate model selection strategies. Moderately complex regional models involving contemporaneous meteorological data from a network of stations were found to be superior to single-site models as well as more complex regional models involving both contemporaneous and time-lagged meteorological data from a network of stations.