The sources are cited in the text. That data came from Wolfgang Schmidt and Petr Necas' respective books:
Schmidt, W. 2001. Chamaeleo calyptratus: The Yemen Chameleon. Matthias Schmidt Publications.
Necas, P. 2004. Chameleons: Nature's Hidden Jewels. 2nd Edition. Edition Chimaera.
As I indicated, these sources note that this area can receive as much as 2,000mm of rainfall a year, although I suspect this is a note of the higher end of the range experienced and not an annual average. I have seen additional sources noting that the wettest areas do receive over 1,000mm annually, however (WWF).
My understanding from multiple sources is that the western slopes of the western highlands is rain belt of the region and each of these sources describe these areas as quite lush. Obviously we are not talking about rainforest "lush", but "lush" for the Arabian peninsula. Further, these sources all indicate that the western slopes experience heavy rainfall as coastal clouds are pushed up in elevation by the wind. Given the elevational change acting as the driver of the rainfall, nearby areas outside of this specific habitat are likely to experience considerably different conditions, which is possibly why the areas you visited did not experience nearly as much rainfall.
Further, water
flowing year round is not the same as it being
available. As you said, wadis are dried, except in the rainy season, however they continue to contain vegetation and the microhabitat this creates results in an increased humidity and even water availability for the chameleons in the form of overnight and morning dew due to the cooling temperatures overnight.
Unfortunately this map is not entirely accurate. The authors of this assessment (
http://www.iucnredlist.org/details/176306/0) have mistakenly limited the distribution of this species to elevations between 1,200 and 2,000 m asl, while the nominate subspecies ranges from 500 and 2850m asl (Schmidt, 2001; Tilbury, 2010) and the subspecies
C. c. calcarifer ranges from the coastal plains up to 1200m asl in its range (Tilbury, 2010). While the eastern boundaries are relatively accurate, this elevation difference would result in a fair westward increase in the distribution shown in this map.
As described above, the true range of this species taken into account, this is actually more easterly in the range of
C. calyptratus. Further, as you travel toward Yarim from Ibb, you are traveling into the heart of the central high plains, which is much drier, and where the species is more restricted and supposedly less common.
Chris