Download Rainy Days V UPDATED

At the studio now we make Americanos and pull up our rainy day playlist, the sounds that echo through Ann Street Studio on days just like today in New York. The music fills my soul when I look out the window at the beautiful simplicity of water falling and washing away all our yesterdays.

I love rainy days. I love the sound. I love the smell. When it's overcast, I light candles. Rainy Days put me in the mood to stay home on the couch and curl up with a good book. To me, this color creates that same feeling.

Download Rainy Days V

Download File 🔥 https://urluss.com/2y3KJP 🔥

Future changes in the structure of daily rainfall, especially the number of rainy days and the intensity of extreme events, are likely to induce major impacts on rain-fed agriculture in the tropics. In Africa this issue is of primary importance, but the agreement between climate models to simulate such descriptors of rainfall is generally poor. Here, we show that the climate models used for the fifth assessment report of IPCC simulate a marked decrease in the number of rainy days, together with a strong increase in the rainfall amounts during the 1% wettest days, by the end of the 21st century over Southern Africa. These combined changes lead to an apparent stability of seasonal totals, but are likely to alter the quality of the rainy season. These evolutions are due to the superposition of slowly-changing moisture fluxes, mainly supported by increased hygrometric capacity associated with global warming, and unchanged short-term atmospheric configurations in which extreme events are embedded. This could cause enhanced floods or droughts, stronger soil erosion and nutriment loss, questioning the sustainability of food security for the 300 million people currently living in Africa south of the Equator.

Figure 3 generalizes the analysis of rainfall evolutions over the Malawi and Tanganyika sectors defined in Fig. 2, for the whole distribution of daily rain totals, and not only the rainy days and the extreme events. Given that the climate models still simulate biased distributions of daily rainfall amounts, especially in the tropics, we compare here the structure of rainfall in the historical simulations (HIST, representative of the recent decades under observed concentrations of greenhouse gases) and the future distributions obtained from the emission scenarios RCP2.6 and RCP8.5. Attention is paid to the agreement between climate models, to quantify the probability of occurrence of each simulated evolution.

In the Tanganyika region, changes in rainfall distribution are more sensitive to the greenhouse gas emission scenario. There is a significant increase in heavy rainfall days associated with RCP8.5 (Fig. 3), while other rainfall categories exhibit weaker, yet significant, changes. Results are also highly reproducible by most climate models. The changes there lead to wetter conditions by the end of the century, suggesting rather favorable conditions for rain-fed agriculture. Yet, this increase occurring mostly during a limited number of intense events, associated impacts for local societies could be weaker than expected, or even detrimental for agronomic yields.

During the p99 events over the Malawi sector more specifically, similar long-term changes are noted, although the spatial patterns are noisier due to smaller sample sizes (Fig. 4b). Stronger moisture convergence prevails over tropical Southern Africa during future extreme rainfall events. This suggests that the average changes discussed above will be exaggerated during heavy rainfall events. Figure 4c further investigates this issue. Instead of computing the differences between future and present p99 events (as in Fig. 4b), we analyze here the short-lived anomalies against their corresponding 30-year climatology (that is, the climate mean state for the early, mid- or late century). From one period to another, including both historical and RCP8.5 simulations, short-term moisture flux and convergence anomalies during heavy rainfall days are remarkably stable (Fig. 4c). Lower-layer fluxes converge over tropical Africa and convey precipitable there, originating from both Atlantic and Indian Ocean basins. Southeasterly anomalies over the south-west Indian Ocean denote an abnormally weak South Indian Convergence Zone19 during these days. This promotes meridional convergence over the subcontinent. Moisture divergence anomalies prevail on both sides of the Great Lakes, on the South Atlantic and along the east coast of Southern Africa. The stationarity of these features indicate that the increase in the p99 values (Fig. 2) results from the superposition of short-lived (synoptic-scale) circulation anomalies, which are mostly unmodified from the late 20th century until the end of the 21st century, and slowly-varying long-term changes directly driven by anthropogenic greenhouse gas emissions. The latter are therefore strongly sensitive to the emission scenarios for future decades, the most optimistic scenario RCP2.6 corresponding to the lowest increase in the extreme event intensities (Figs 2e and 3).

These results show that the climate models are much more consistent for projecting changes in the structure of daily rainfall, namely more intense extreme events and fewer rainfall days2,10, than for seasonal totals. More precise and detailed results could be obtained from high-resolution limited area models, shown to outperform global models, especially for simulating intense rainfall events20. These changes simulated over Southern Africa may sensibly alter the quality of the future rainy seasons, from an agronomic perspective. More frequent dry days or more persistent dry sequences may cause strong decrease in agronomic yields, if occurring during key phases of the phenology (e.g., the flowering phase) when the plant is vulnerable15. Abundant rainfall amounts may also reduce soil fertility due to nitrogen leaching. Although this could add uncertainties and make the modeling chain more complex, inter-comparing crop-models over regions experiencing such evolutions in their daily rainfall (see for instance )21 could allow separating the impacts of climate change from the errors and uncertainties associated with both agronomic and climate models. This could help anticipating such changes and adopt relevant mitigation strategies.

How to cite this article: Pohl, B. et al. Fewer rainy days and more extreme rainfall by the end of the century in Southern Africa. Sci. Rep. 7, 46466; doi: 10.1038/srep46466 (2017).

Rainfall days from ERA-Interim, ERA5, and MERRA-2 reanalyses between 1980 and 2016. (a) Average number of rainfall days per year, (b) interannual variability in the number of rainfall days per year, and (c) trend in the number of rainfall days per year.

Ratio of rainfall days to total precipitation days for ERA-Interim, ERA5, and MERRA-2 for the entire Arctic and regions. Standard deviations are in parentheses. Trends are in percent per decade. Trends that are statistically significant at the 90th percentile are in bold.

The increasing number of above-freezing days in the Arctic might be tied to increasing rainfall frequency. Figure 7b shows the spatial correlations between the number of days above freezing and the number of rainfall days. MERRA-2 has the highest spatial correlations overall, specifically in the peripheral seas and the central Arctic. All reanalyses show high correlations in the North Atlantic. Figure 8 presents scatterplots of the number of days where the temperature is above freezing and the number of rainfall occurrences, for the entire Arctic and specific regions, and demonstrates that these variables, although not one-to-one, are highly correlated. While there are many more days above freezing than rainfall occurrences, the more often above-freezing days occur, the higher the likelihood of rainfall (Fig. 8a). Regionally, the highest correlations between temperature and rainfall occur in the North Atlantic, followed by the peripheral seas, and the lowest correlations in the central Arctic (Figs. 8c,d). ERA-Interim (MERRA-2) has the lowest (highest) correlations overall.

When comparing the first day above freezing and the first rainfall day, there are high correlations in the North Atlantic and the peripheral seas near the coasts, with much smaller and slightly negative correlations present in the central Arctic (Fig. 11a). ERA-Interim has many occurrences when the first rainfall day occurs much earlier than the first day above freezing (Fig. 11b). MERRA-2 shows some instances of earlier rainfall, while ERA5 has rainfall occurring nearly always after the first day above freezing. When split into the individual regions (Fig. 11c), it is apparent that ERA-Interim produces rainfall before warm temperatures in all regions, especially in the central Arctic, where rainfall is occurring around day 50 (19 February) but the first day above freezing is consistent around day 160 (9 June). These scatterplots also highlight a delay or lag in ERA5 and MERRA-2 between the first day above freezing and first day of rainfall (points below the one-to-one line) (Figs. 11b,c), and not for ERA-Interim. This is driven by the fact that just because average daily temperatures reach the freezing point does not guarantee that is going to precipitate that day, just that if it were to precipitate, then atmospheric temperatures could be warm enough for that precipitation to be rainfall. This lag for ERA5 and MERRA-2 tends to be around 20 days. For MERRA-2, and to a greater extent for ERA-Interim, in the central Arctic, the first rainfall day sometimes occurs before the first warm day of the year (points above the one-to-one line) and could be occurring when temperatures are hovering just below freezing. This occurs to a much lesser extent in ERA5. These instances may occur when the temperature was above freezing for a brief period of the day when rainfall occurred, but the average daily temperature could have been below freezing due to the diurnal cycle that is more extreme in the spring and fall months. 2351a5e196