South-eastern Spain is a key area for assessing the effects of climate change on biodiversity since it presents an ecotone between the Mediterranean biome and the subtropical shrublands of arid lands. The forests of Tetraclinis articulata constitutes an especially relevant case. A species distribution model has been developed, regionalised climate change scenarios for South-eastern Spain were generated and expected changes in the suitability area of this species were estimated under B2 and A2 SRES scenarios for the time slice 2020-2050. Moreover, land use in the present and future potential habitat has been analysed. The high sensitivity of T. articulata is expressed not only as effects of climate change in the near future when compared to the present-day situation but also in the remarkable differences under scenarios B2 and A2. Under scenario B2 the suitable area for T. articulata would expand six-fold whereas under A2 the potential habitat would disappear from its present-day distribution and would move to a small area in the interior mountains. Under scenario B2 the future potential habitat in the coastal location would include enough area of shrublands, the main effective habitat of the species. Moreover, the present and future potential habitat partially overlaps, which facilitates the species migration. On the contrary, in the interior potential habitat the land use is less favourable for the effective habitat, the actual and future potential habitat do not overlap and the low dispersal capabilities of the species prevents natural migration to the interior to be expected.

No abstract

Real-word errors are characterized by being actual terms in the dictionary. By providing context, real-word errors are detected. Traditional methods to detect and correct such errors are mostly based on counting the frequency of short word sequences in a corpus. Then, the probability of a word being a real-word error is computed. On the other hand, state-of-the-art approaches make use of deep learning models to learn context by extracting semantic features from text. In this work, a deep learning model were implemented for correcting real-word errors in clinical text. Specifically, a Seq2seq Neural Machine Translation Model mapped erroneous sentences to correct them. For that, different types of error were generated in correct sentences by using rules. Different Seq2seq models were trained and evaluated on two corpora: the Wikicorpus and a collection of three clinical datasets. The medicine corpus was much smaller than the Wikicorpus due to privacy issues when dealing with patient information. Moreover, GloVe and Word2Vec pretrained word embeddings were used to study their performance. Despite the medicine corpus being much smaller than the Wikicorpus, Seq2seq models trained on the medicine corpus performed better than those models trained on the Wikicorpus. Nevertheless, a larger amount of clinical text is required to improve the results.