![Hier klicken, um den Treffer aus der Auswahl zu entfernen](images/unchecked.gif) |
Titel |
Validation and sensitivity analyses of a soil temperature and moisture model under pedoclimatic conditions of southern Quebec (Canada) |
VerfasserIn |
Simon Perreault, Karem Chokmani, Michel Nolin, Gaétan Bourgeois |
Konferenz |
EGU General Assembly 2011
|
Medientyp |
Artikel
|
Sprache |
Englisch
|
Digitales Dokument |
PDF |
Erschienen |
In: GRA - Volume 13 (2011) |
Datensatznummer |
250047821
|
|
|
|
Zusammenfassung |
Soil moisture and temperature conditions play an important role in plant growth and
development. Modelling soil moisture and temperature could allow predicting yield and
outbreaks of pest insects, diseases, asphyxia, and drought risks. In this study, the Soil
Temperature and Moisture Model (STM2), developed by USDA, was used to generate soil
moisture and temperature predictions at different depths: 15, 30, 45, and 60 cm
for moisture and 10 cm for temperature. This model has two kinds of inputs: 1)
soil properties (soil texture and organic matter content) and 2) daily weather data
(total rainfall, minimum and maximum air temperatures). During the 2008 and
2010 growing seasons (mid-May to mid-October), soil moisture and temperature
measurements were obtained from three pedoclimatic stations equipped with EC-10 soil
moisture probes and HOBO soil temperature probes. Those measurements have been
used as references to validate the STM2 predictions. These pedoclimatic stations
were located in three corn fields of southern Quebec (Canada) representing the soil
surface texture diversity observed in this agricultural area: loamy sand (LS), sandy
loam (SL) and silty clay (SiC). For each pedoclimatic station, a soil profile was
sampled and analyzed to measure soil texture (coarse fragments (>2 mm), sand, silt
and clay percentages), organic matter content, bulk density, saturated hydraulic
conductivity (Ksat), soil moisture content at different tensions (permanent wilting point at
1500 kPa, field capacity at 33 kPa, and saturation at 0 kPa). Weather data were
collected at Environment Canada weather stations located in the vicinity of corn fields.
Relative sensitivity analyses were conducted on the STM2 pedotransfer functions
(PTF) in order to determine which soil properties has the greatest impact on soil
moisture and temperature predictions. The model performance was evaluated using the
modelling index d and two error measurements (RMSE = Root Mean Square Error and
Bias) computed for the global growing season and at three corn growth periods: 1)
seeding to emergence, 2) emergence to flowering and 3) flowering to senescence. The
global performance of soil temperature predictions was better than soil moisture
predictions. The estimation quality decreases with increasing depth and is higher
during the first and third growth periods. Daily predictions were better than hourly
predictions for soil moisture but they were mostly the same for soil temperature. Good
performances were observed for LS and SiC. The model performance was slightly lower
for the SL soil. The PTF used in the STM2 were mostly satisfactory. However,
the PTF used for estimating Ksat showed lower performance. Nevertheless, the
STM2 sensitivity analyses revealed that Ksat plays a marginal role in soil moisture
and temperature predictions. In fact, the STM2 is more sensitive to bulk density.
The PTF sensitivity analyses demonstrated that permanent wilting point was more
affected by clay percentage, field capacity by clay and sand percentages, saturated
hydraulic conductivity by clay and silt percentages, and bulk density by organic
matter and clay percentages. Therefore, this preliminary study shows that the STM2
model could be used in combination with soil and climatic data sets for reliably
predicting soil moisture and temperature variations in southern Quebec (Canada). |
|
|
|
|
|