Читать книгу Congo Basin Hydrology, Climate, and Biogeochemistry - Группа авторов - Страница 30

There is an increasing number of recent studies using regional and global models to assess climate over central Africa (e.g., Aloysius et al.; 2016; Creese and Washington 2016, 2018; Dosio et al., 2019; Fotso‐Kamga et al., 2019; Haesnler et al., 2013; Sonkoué et al., 2018; Taguela et al., 2020; Tamoffo et al., 2019; Tchotchou and Mkankam, 2010; Vondou and Haensler 2017; Washington et al., 2013). Using CMIP5 global models, Aloysius et al. (2016) reveal that skills of simulated temperature is better than those of rainfall. There is an important discrepancy in the climatology of rainfall appearing in the seasonality, spatial patterns, and magnitude of precipitation. Tamoffo et al. (2020) highlight the importance of monitoring moisture variables and strength of low‐level flow that transports moisture toward the central African region. The findings of a large ensemble of climate models convey dissimilarity but possible outlines for the rainfall change are owing to the contrasts of climatology features across models. In their investigation, Creese and Washington (2016) demonstrated that simulated precipitation depends on the penetration of the moisture in the Congo Basin, where CMIP5 models strongly disagree. They also call for reinforcement of observations for a better description of the processes interacting and also required to represent convection explicitly in models in the Congo Basin.

Taguela et al. (2020) show that models involved in the CORDEX‐Africa project are able to capture basic features of temperature and precipitation despite the persistence of local biases. In contrast to what is obtained in other African regions, the model ensemble mean does not always depict better results than individual simulation. This interesting fact highlights the reliability of model output, as already mentioned by Washington et al. (2013). Vondou and Haensler (2017) show that increasing model resolution is not always the way to proceed for the enhancement of climate simulation over central Africa, consistent with analysis of Wu et al. (2019). It was pointed out that model formulation mostly impacts precipitation production rather than refinement of resolution. However it is important to note that the amplitude of the diurnal cycle of precipitation is influenced by spatial resolution of the model. For the improvement of the regional model RCA4 to mimic precipitation over the Congo Basin, Tamoffo et al. (2019) use an ensemble of experiments based on moisture transport and regional circulation analysis. Henceforth, they raise the question of the credence of models over central Africa and highlight the fact that much effort should be given to the improvement of boundary layer representation and land‐atmosphere coupling. Another big problem models encounter in central Africa is representation of land–sea contrast. Vondou and Haensler (2017), using regional climate REMO, show that the model tends to overestimate rainfall over the ocean while simulating less precipitation over the Congo Basin in accordance with recent results of Fotso‐Kamga et al. (2019), when they analyzed COSMO model output. Misunderstanding of the central African climate engine can lead to poor mimicking of key regional processes, discrepancies in simulated precipitation, and in turn important unreliable information in the guidelines for the future climate (Cook & Vizy, 2019; Creese et al., 2019; Tamoffo et al., 2020).