A Study of the Relations between Soil Moisture, Soil Temperatures and Surface Temperatures Using ARM Observations and Offline CLM4 Simulations

Climate

Menglin S. Jin¹  Terrence Mullens² 

[1]Department of Meteorology and Climate Science, San José State University, San José, CA 95192, USA

[2] E-Mail

关键词: land surface climatology;    climate modeling;    soil moisture;    surface temperature;    land surface model;    ground observations;   

DOI  :  10.3390/cli2040279

来源: mdpi

【 摘 要 】

Soil temperature, soil moisture, skin temperature and 2-m air temperature are examined from both ground observations and the offline community land model (CLM4). Two-layer soil moisture and three-layer soil temperature observations from six-year (2003–2008) ground measurements at the Lamont, Oklahoma site supported by the Atmospheric Radiation Measurement (ARM) Program of the Department of Energy (DOE) show clear vertical and temporal relations between soil temperature and soil moisture with surface skin temperature and 2-m air temperature. First, daily means reveal that all of these variables have clear seasonal variations, with temperatures peaking in summer and minimizing in winter as a result of surface insolation. Nevertheless, the 2-m air temperature and upper soil temperature (−0.05 m) peak at 2 h after that of surface skin temperature because of the lag of transport of heat from the skin level to the 2-m air and to underground respectively. As a result of such lag, at the monthly annual cycle scale, 2-m air temperature has higher correlation with upper soil temperature than skin temperature does. Second, there are little diurnal and annual variations at the lowest soil layer (−0.25 m). Third, a negative correlation (~−0.40) between skin temperature and soil moisture is observed, consistent with the expectation that heat flux and evaporation are competing physical processes for redistributing surface net radiation. Soil moisture, however, minimizes in March and maximizes in winter due to the local rainfall cycle. All of these key observed relations are qualitatively reproduced in the offline CLM4 using the atmosphere forcing derived from ARM observations. Nevertheless, CLM4 is too dry at the upper layer and has less variation at the lower layer than observed. In addition, CLM4 shows stronger correlation between T_soil and T_skin (r = 0.96) than the observations (r = 0.64), while the predicted nighttime T_skin is 0.5–2 °C higher than the observations.

【 授权许可】

CC BY   
© 2014 by the authors; licensee MDPI, Basel, Switzerland.