Abstract
There is necessity of considering air temperature to simulate the hydrology and management within water resources systems. In many cases, a big issue is considering the scarcity of data due to poor accessibility and limited funds. This paper proposes a methodology to obtain high resolution air temperature fields by combining scarce point measurements with elevation data and land surface temperature (LST) data from remote sensing. The available station data (SNOTEL stations) are sparse at Rocky Mountain National Park, necessitating the inclusion of correlated and well-sampled variables to assess the spatial variability of air temperature. Different geostatistical approaches and weighted solutions thereof were employed to obtain air temperature fields. These estimates were compared with two relatively direct solutions, the LST (MODIS) and a lapse rate-based interpolation technique. The methodology was evaluated using data from different seasons. The performance of the techniques was assessed through a cross validation experiment. In both cases, the weighted kriging with external drift solution (considering LST and elevation) showed the best results, with a mean squared error of 3.7 and 3.6 °C2 for the application and validation, respectively.
Highlights
Air temperature is a key variable in the hydrological cycle
We propose a novel approach to obtain an optimal estimation of temperature fields by combining scarce point measurements from SNOTEL stations, land surface temperature (LST) from MODIS, and elevation from a digital elevation model (DEM)
The scarce air temperature data from SNOTEL were complemented by well-sampled secondary information from elevation and LST in order to obtain air temperature fields
Summary
Air temperature is a key variable in the hydrological cycle. It is a driver of processes such as evaporation [1], sublimation [2], and snow melt [3]. Distributed temperature data are often required, especially in complex terrain where the spatial variability is large, and data are sparse. Air temperature is influenced by several land and atmospheric processes and can have an important variability due to spatiotemporal changes in these processes [7]. Depending on the scale of interest, the air temperature patterns can be very different for the same case study [8]. Air temperature data are usually scarce, especially in montane regions, due to the problems of maintaining the monitoring system under harsh climatic conditions [9].
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call