Identifier

etd-04032012-091137

Degree

Doctor of Philosophy (PhD)

Department

Plant, Environmental Management and Soil Sciences

Document Type

Dissertation

Abstract

The Transylvanian Plain (TP) is a 395,000 ha region located in north-central Romania and is an area of agronomic importance in the region. The TP is characterized by hilly terrain, dissected by the Someş and Mureş Rivers. The terrain creates a unique situation when assessing pedology and soil temperature. Soils can change quickly across the landscape in the TP due to the terrain. To account for these differences, soil temperature was measured to predict soil temperatures as well as to evaluate growing conditions. Twenty stations were installed for a long-term temperature and pedology study. Pedons were described for morphological characterization at each location. Pedon descriptions were then classified using both US Soil Taxonomy (USST) and Sistemul Roman De Taxonomie A Solurilor (Romanian System of Soil Taxonomy- RSST). The two soil classification systems aligned for all 20 stations. Morphological descriptions showed that there were 10 Mollisols (Cernisoluri), 4 Alfisols (Luvisoluri), and 6 Inceptisols (Cambisoluri) according to USST (RSST). All locations had sufficient organic carbon to classify as mollic epipedons. However, other requirements such as: color and depth of epipedon were not met. Soil temperature is identified at the family level in USST and is not present in RSST. In addition to morphological characterization at the 20 locations, soil and air temperatures were measured via a data logging system. Soil temperature is a vital property when evaluating crop growth due to its influence on germination and root growth. Growing degree days (GDD) were evaluated for the summer of 2009 using air temperature for the TP. Craiesti and Filpisu Mare were significantly warmer than Matei and Zoreni and gained sufficient GDD for tasseling 21 days earlier. Mean annual soil temperature (MAST) was predicted using a multiple regression model and Landsat 7 ETM+. Landsat provided a better linear relationship to in situ MAST values with a coefficient of determination value (R2) of 0.63 compared to the multiple regression with an R2 of 0.42. Significant differences were found in MAST values between agricultural and urban land covers. The use of Landsat ETM+ could reduce the time and expense of large in situ field studies.

Date

2012

Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Weindorf, David C.

DOI

10.31390/gradschool_dissertations.2145

Share

COinS