Dataset for "Improving Thermal Comfort in Refugee Shelters in Desert Environments"

The dataset describes monitored environmental conditions of unoccupied shelter prototypes in the refugee camp of Azraq (Jordan). The monitored environmental conditions are temperature and relative humidity every hour both outdoors and indoors. The 12 shelter prototypes include a control shelter without modifications and 11 variants implementing a range of passive/active measures (increased ventilation, insulation, thermal mass and/or roof shades, evaporative cooler and earth tube).

refugee shelters, thermal comfort, thermal modeling, overheating, desert
Civil engineering and built environment
Climate and climate change

Cite this dataset as:
Moran, F., Fosas de Pando, D., Coley, D., Natarajan, S., Orr, J., 2021. Dataset for "Improving Thermal Comfort in Refugee Shelters in Desert Environments". Bath: University of Bath Research Data Archive. Available from:


[QR code for this page]


application/zip (535kB)
Creative Commons: Attribution 4.0

Raw and clean data

Data WS-GP1 Azraq
application/zip (1MB)
Creative Commons: Attribution 4.0


Francis Moran
University of Bath

David Coley
University of Bath

John Orr
University of Cambridge


Collection date(s):

From 15 July 2019 to 21 September 2019

Geographical coverage:

Jordan, Zarqa Governorate, Refugee Camp of Azraq


Data collection method:

The work was completed using two hypotheses: 1. It is possible to take an existing shelter design, that is already deployed at scale and has been shown via field surveys to provide unsatisfactory thermal comfort, and complete a parametric analysis using field testing of prototypes to generate an affordable derivative of the same appearance that provides improved internal thermal conditions as measured against acceptable thermal comfort bands. 2. That similar results can be achieved using thermal modelling without any field testing even though many of the modelling inputs are unlikely to be accurately known. To assess our hypotheses, the thermal conditions within eleven variants of identical form but representing different possible improvement strategies were compared to a base design. The success of the prototypes was assessed by comparing the maximum daytime temperature and number of degree hours over comfort levels compared to the control shelter. The success of computational modelling was to be assessed by comparing predicted performance against monitored performance. Twelve unoccupied test shelters were monitored (temperature, humidity, opening of doors and windows) for ten weeks from 15/7/19 – 21/9/19. Nine shelters received passive adaptations, two received active adaptations, all adaptations were focused on improving internal summer thermal comfort. The shelters were simulated in EnergyPlus v9.0.1 Two metrics are used to evaluate the thermal performance of each shelter: air temperature and operative temperature. Overheating is typically quantified as the total number of hours the threshold has been surpassed (total hour count, OH_ch [h]). However, such a metric fails to account for the severity of overheating, since deviations of 1 K over the threshold are more benign than deviations of 6 K. For this purpose, a second metric weighs (multiplies) the duration of overheating according to the temperature deviation above the threshold (to give weighted hours of overheating, OH_wh [K·h]). Here we use the ASHRAE Standard 55 Thermal Comfort Model together with that found in Vellei. The change in internal conditions provided by the interventions are assessed for both the physical and modelled shelters using the following metrics: 1. Minimum air temperature difference between indoors and outdoors, rTp-e (°C). Negative numbers indicate the extent to which air temperature in the shelter is cooler than the external air temperature. 2. Maximum indoor air temperature difference between shelter variant and control shelter, rTp-c (°C). Negative numbers indicate the extent to which the shelter is cooler than the control shelter. 3. Minimum indoor air temperature, T_min (°C) 4. Mean indoor air temperature, T_avg (°C) 5. Maximum indoor air temperature, T_max (°C) 6. T_comf_max_ashrae: ASHRAE upper limit for indoor operative temperature (80% acceptability) (°C) 7. T_comf_max_vellei: Vellei’s model upper limit for indoor operative temperature (80% acceptability) (°C) 8. Total overheating hours compared with the adaptive comfort temperature, OH_ch (h). The metric is provided for the two thresholds considered, ASHRAE’s and Vellei’s. 9. Total overheating Kelvin hours compared with the adaptive comfort upper temperature, OH_wh (K·h). The metric is provided for the two thresholds considered, ASHRAE’s and Vellei’s.

Documentation Files

Readme full text.docx
application/vnd.openxmlformats-officedocument.wordprocessingml.document (19kB)
Creative Commons: Attribution 4.0

text/plain (3kB)
Creative Commons: Attribution 4.0


Engineering and Physical Sciences Research Council (EPSRC)

Healthy Housing for the Displaced

Publication details

Publication date: 20 January 2021
by: University of Bath

Version: 1


URL for this record:

Related articles

Moran, F., Fosas, D., Coley, D., Natarajan, S., Orr, J. and Ahmad, O. B., 2021. Improving thermal comfort in refugee shelters in desert environments. Energy for Sustainable Development, 61, pp.28-45. Available from:

Contact information

Please contact the Research Data Service in the first instance for all matters concerning this item.

Contact person: Francis Moran


Faculty of Engineering & Design
Architecture & Civil Engineering

Research Centres & Institutes
Building Research Park
Centre for Doctoral Training in Decarbonisation of the Built Environment (dCarb)
Centre for Energy and the Design of Environments (EDEn)