Estimation of Relative Humidity and Dew Point Temperature Using Limited Meteorological Data

Hydrological processes such as evaporation and evapotranspiration (ET) are affected by air humidity. Air humidity is commonly expressed as relative humidity (RH) or dew point temperature (Tdew) and is required for the determination of evaporation from water or soil surfaces and ET from cropped areas. However, the availability of humidity data is sometimes an issue. In cases in which RH data are not available, air humidity on a daily basis can be estimated by using daily maximum and minimum temperatures along with the dew point temperature. Guidelines from an international organization recommend the use of a dew point temperature slightly less than the daily minimum temperature for daily computations, when its measurements are not available. In this study, the daily average dew point temperature (Tdewmean) was taken as k°C less than the daily minimum temperature. Thus, daily RH data (maximum, minimum, and mean RH) were generated from daily maximum and minimum temperatures for Roorkee Station (India) during 2012–2016. Four different values of k (k=0, 1, 2, 3) were used for determining the actual vapor pressure values used in RH estimation. Daily RH values determined using different values of k were compared with observed RH data. It was observed that for the nonsummer months, having an average value of daily mean RH greater than 60% (January–March and July–December), k=0 resulted in minimum error in RH estimation. However, during the summer months, having an average value of daily mean RH less than 60% (April–June), k=3 provided relatively better estimates of RH. During the summer months, large errors occurred in the estimation of maximum and minimum RH values using this approach; hence, mean RH values that are relatively better estimated during these periods should be used for hydrological applications. Further, linear Tdew-RH relationships (with and without parabolic correction) have been assessed to determine daily Tdewmean values. These relationships have been found to provide more accurate estimates of Tdewmean and actual vapor pressure on a daily basis than the FAO-56 vapor pressure models when only daily RHmean data are available (instead of RHmax and RHmin data) along with daily temperature data.