Robert Walko

Large-scale semi-arid afforestation can enhance precipitation and carbon sequestration potential

Afforestation is an important approach to mitigate global warming. Its complex interactions with the climate system, however, makes it controversial. Afforestation is expected to be effective in the tropics where biogeochemical and biogeophysical effects act in concert; however, its potential in the large semi-arid regions remains insufficiently explored. Here, we use a Global Climate Model to provide a process-based demonstration that implementing measured characteristics of a successful semi-arid afforestation system (2000 ha, ~300 mm mean annual precipitation) over large areas (~200 million ha) of similar precipitation levels in the Sahel and North Australia leads to the weakening and shifting of regional low-level jets, enhancing moisture penetration and precipitation (+0.8 ± 0.1 mm d−1 over the Sahel and +0.4 ± 0.1 mm d−1 over North Australia), influencing areas larger than the original afforestation. These effects are associated with increasing root depth and surface roughness and with decreasing albedo. This results in enhanced evapotranspiration, surface cooling and the modification of the latitudinal temperature gradient. It is estimated that the carbon sequestration potential of such large-scale semi-arid afforestation can be on the order of ~10% of the global carbon sink of the land biosphere and would overwhelm any biogeophysical warming effects within ~6 years.

Improving Atmospheric Model Performance on a Multi-Core Cluster System

Numerical models have been used extensively in the last decades to understand and predict weather phenomena and the climate. In general, models are classified according to their operation domain: global (entire Earth) and regional (country, state, etc). Global models have spatial resolution of about 0.2 to 1.5 degrees of latitude and therefore cannot represent very well the scale of regional weather phenomena. Their main limitation is computing power. On the other hand, regional models have higher resolution but are restricted to limited area domains. Forecasting on limited domain demands the knowledge of future atmospheric conditions at domain’s borders. Therefore, regional models require previous execution of global models.

Sensitivity Studies on the Impact of Dust and Aerosol Pollution Acting as Cloud Nucleating Aerosol on Orographic Precipitation in the Colorado River Basin

In this study, we examine the cumulative effect of pollution aerosol and dust acting as cloud nucleating aerosol;cloud condensation nuclei (CCN), giant cloud condensation nuclei, and ice nuclei (IN), on orographic precipitation in the Rocky Mountains. We analyze the results of sensitivity studies for specific cases in 2004-2005 winter season to analyze the relative impact of aerosol pollution and dust acting as CCN and IN on precipitation in the Colorado River Basin. Dust is varied from 3 to 10 times in the experiments, and the response is found to be nonmonotonic and depends on various environmental factors. The sensitivity studies show that adding dust in a wet system increases precipitation when IN effects are dominant. For a relatively dry system high concentrations of dust can result in overseeding the clouds and reductions in precipitation. However, when adding dust to a system with warmer cloud bases where drizzle formation is active, the response is nonmonotonic.

Atmospheric models hybrid OpenMP/MPI implementation multicore cluster evaluation

Atmospheric models usually demand high processing power and generate large amounts of data. As the degree of parallelism grows, the I/O operations may become the major impacting factor of their performance. This work shows that a hybrid MPI/OpenMP implementation can improve the performance of the atmospheric model ocean-land-atmosphere model (OLAM) on a multicore cluster environment. We show that the hybrid MPI/OpenMP version of OLAM decreases the number of output files, resulting in better performance for I/O operations. We have evaluated OLAM on the parallel file system PVFS and shown that storing the files on PVFS results in lower performance than using the local disks of the cluster nodes due as a consequence of file creation and network concurrency. We have also shown that further parallel optimisations should be included in the hybrid version in order to improve the parallel execution time of OLAM.