We review the recent progress in dynamical and statistical downscaling approaches for west African precipitation and perform a regional climate model (RCM) intercomparison using the novel multi-model RCM data set from the Ensembles-based Predictions of Climate Changes and Their Impacts (ENSEMBLES) and African Monsoon Multidisciplinary Analyses (AMMA) projects. Present RCMs have distinct systematic errors in terms of west African precipitation varying in amplitude and pattern across models. This is also reflected in a relatively large spread in projected future precipitation trends. Altogether, the ENSEMBLES RCMs indicate a prevailing drying tendency in sub-Saharan Africa. Statistical post-processing of simulated precipitation is a promising tool to reduce systematic model errors before application in impact studies.

Single ozonation of two nitroaromatic hydrocarbons (nitrobenzene and 2,6-dinitrotoluene) under different experimental conditions (ozone feed rate, pH, temperature, hydroxyl radical scavengers) has been studied. The absence of hydroxyl radical scavengers, pHs 7−9, and temperatures below 30 °C are optimum conditions for nitroaromatic removal. Due to the importance of hydroxyl radical reactions, removal rates in natural water are much lower than those observed in laboratory ultrapure water. Rate constants of the direct reaction between ozone and nitroaromatic hydrocarbons at 20 °C have been found to be lower than 6 M-1 s-1. More than 99% of nitroaromatic removal is due to hydroxyl radical oxidation. Single ozonation of nitroaromatics can then be classified as a real advanced oxidation technology. Nitrophenols, compounds very reactive toward ozone and hydroxyl radicals, and 2,6-dinitrobenzaldehyde, identified in the single ozonation of nitrobenzene and 2,6-dinitrotoluene, respectively, are some of the first intermediates of single ozonation.

A kinetic model for the advanced oxidation (ozonation alone, UV radiation alone, ozone plus hydrogen peroxide, ozone plus UV radiation, and UV radiation plus hydrogen peroxide) of aromatic hydrocarbons in water is proposed and tested with experimental results of the oxidation of nitrobenzene and phenanthrene, two aromatic hydrocarbons of different reactivity with ozone. The kinetic model leads to good results in the case that the compound treated reacts exclusively with ozone, that is, without the contribution of hydroxyl radical oxidation as in the case of phenanthrene oxidation. In this case, it is not necessary to account for intermediate reactions to have good predictions of experimental remaining concentrations of ozonation processes. On the contrary, when the aromatic hydrocarbon is mainly removed by hydroxyl radicals (case of nitrobenzene), mole balance equations of intermediates have to be included for the experimental concentrations to be reproduced. For so doing, the kinetic parameters, such as rate constants of reactions between ozone and hydroxyl radical with intermediates and their corresponding quantum yields at 254 nm, were also determined. The kinetic model, however, is unable to reproduce, with accuracy, the experimental results of the ozone−UV radiation oxidation system.

The combined ozonation of nitrobenzene and 2,6-dinitrotoluene with hydrogen peroxide and UV radiation has been studied. Effects of variables such as pH, hydroxyl radical scavengers, ozone feed rate, or hydrogen peroxide to ozone feed molar ratios have been investigated. Results are also compared to those obtained from UV radiation and single ozonation (Beltrán et al., 1998). Quantum yields of nitrobenzene and 2,6-dinitrotoluene at 254 nm were found to be 7.0 × 10-3 and 2.2 × 10-2 mol einstein-1, respectively. Removal rates of nitrobenzene through single or combined ozonation with hydrogen peroxide at concentration lower than 10-2 M are similar, while under the same conditions combined ozonation of 2,6-dinitrotoluene leads to higher removal rates than single ozonation. Also, the latter compound is removed faster by UV radiation alone and UV/O3 oxidation than by single ozonation. Rate constants of the reaction between the hydroxyl radical and nitrobenzene and 2,6-dinitrotoluene were found to be 2.9 × 109 and 7.5 × 108 M-1 s-1, respectively. Nitrophenols and 2,6-dinitrobenzaldehyde appear to be some of the principal intermediates formed in the treatment of nitrobenzene and 2,6-dinitrotoluene, respectively, regardless of oxidation techniques applied. Apart from economic considerations UV/O3 oxidation is the technique among those studied leading to the highest removal rates.