Versatile Soil Moisture Budget Model Research Articles

Improved characterization of frozen soil processes in the Versatile Soil Moisture Budget model

Mohammed, G. A., Hayashi, M., Farrow, C. R. and Takano, Y. 2013. Improved characterization of frozen soil processes in the Versatile Soil Moisture Budget model. Can. J. Soil Sci. 93: 511-531. Soil freezing and thawing influence the infiltration of rain and snow melt water and subsequent redistribution, runoff generation, and a host of other processes. Accurate characterization of frozen soil processes in hydrological models is important for their use in managing agricultural activities and water resources. The Versatile Soil Moisture Budget (VSMB) is a relatively simple soil water balance model, which has been widely used in Canada for several decades, but its application has primarily been for crop-growing seasons. We have modified the VSMB to include new algorithms for snow accumulation and melt, soil freezing and thawing, and snowmelt infiltration and runoff; and evaluated its performance using field data from a grassland site in Alberta. The new VSMB model simulates snow processes with reasonable accu...

Improved characterization of frozen soil processes in the Versatile Soil Moisture Budget model

Mohammed, G. A., Hayashi, M., Farrow, C. R. and Takano, Y. 2013. Improved characterization of frozen soil processes in the Versatile Soil Moisture Budget model. Can. J. Soil Sci. 93: 511–531. Soil freezing and thawing influence the infiltration of rain and snow melt water and subsequent redistribution, runoff generation, and a host of other processes. Accurate characterization of frozen soil processes in hydrological models is important for their use in managing agricultural activities and water resources. The Versatile Soil Moisture Budget (VSMB) is a relatively simple soil water balance model, which has been widely used in Canada for several decades, but its application has primarily been for crop-growing seasons. We have modified the VSMB to include new algorithms for snow accumulation and melt, soil freezing and thawing, and snowmelt infiltration and runoff; and evaluated its performance using field data from a grassland site in Alberta. The new VSMB model simulates snow processes with reasonable accuracy and predicts the day of thawing within several days of observation. It also estimates the amount of runoff and its inter-annual variability reasonably well, although the model still has limitations in accurately predicting the vertical distribution of water content. Despite these limitations, the model will be useful for estimating the amount of snowmelt runoff that provides the critical water inputs to wetlands and dugouts, and for understanding the effects of landuse variability on these processes.

Application of the Versatile Soil Moisture Budget Model to Estimate Evaporation from Prairie Grassland

Coupled hydrological modelling of land surface and groundwater is an effective tool for water resources management. The Versatile Soil Moisture Budget (VSMB) model is a simple land-surface model that has been widely used in the Canadian prairies to simulate soil moisture conditions of cropland. Its algorithm is suitable for coupling with a groundwater model, but VSMB has not been rigorously tested for grassland as such. The accuracy of simulated grassland evaporation was evaluated against three years of field data collected using the eddy-covariance technique near Calgary, Alberta. The current version of VSMB substantially underestimated evaporation. After modifications on growth stage parameters, radiation scheme, the drying curve function representing the ratio of actual to potential evaporation, and soil-column length, simulated evaporation was well within the error margin of observed values. In the semi-arid environment of the test site, simulated evaporation was relatively insensitive to soil water storage parameters but was sensitive to the drying curve function and soil-column length.

Multivariate analysis of water-related agroclimatic factors limiting spring wheat yields on the Canadian prairies

Water use by spring wheat and soil water contents at meteorological stations on the Canadian prairies were simulated with the Versatile Soil Moisture Budget model for different crop growth stages. Six water-related agroclimatic indices at five growth stages (seeding–emergence, emergence–jointing, jointing–heading, heading–soft dough and soft dough–harvest) and previous non-growing season were correlated to spring wheat yields in the three prairies provinces and in the entire prairie region for the years 1976–2006. Principal component analysis was applied to explore major modes of joint variability in the regional water-related agroclimatic indices. Canonical correlation analysis was employed to further identify joint variability patterns of the water-related indices associated with regional spring wheat yields. Results showed some common features of the effects of the water-related factors at different growth stages: lower-than-normal moisture stress at the jointing–heading stage favoured spring wheat yields in all three provinces. Regional differences were also seen, for example, a slight moisture stress at the heading–soft dough stage could be beneficial to spring wheat yields in Manitoba because of its relatively wetter climate compared to the other two provinces. The results can be used for a better understanding of the effects of water-related agroclimatic conditions at different growth stages on final spring wheat yields on the Canadian prairies, leading to the improvement of crop management. The results can also be used in regional yield forecasting and in the projection of climate change impacts on crop production. This study provided an example of how to quantify crop–climate relationships by the use of statistical multivariate analysis tools.

Quantification and simulation of soil water on grazed fescue watersheds

A 2-year study was conducted at the Agriculture and Agri-Food Canada Stavely Range Substation, Alberta. The objective was to quantify and simulate the soil water status of small grassland watersheds under 3 grazing intensities and 4 topographic positions. The grazing treatments were ungrazed (or control), heavy (2.4 AUM ha−1) and very heavy (4.8 AUM ha−1) grazing and the topographic positions were upperslope, midslope, lowerslope and 5 m away from the collector drain. Moisture readings were taken every 2 weeks between spring and fall using a CPN 503 moisture neutron probe. Readings were taken at the soil surface and at 15-, 25-, 35-, 45- and 55-cm depths. Total annual precipitation in 1998 and 1999 was 648 and 399 mm, respectively. In both years grazing treatments did not affect total soil water in the 0–50 cm (TSW50) depth interval for the upper, middle and lower slope positions, but TSW50 close to the collector drain was significantly (P ≤ 0.05) greater for the heavy grazed compared to the very heavy grazed treatment. Within each grazing treatment, TSW50 differences among slope positions occurred mainly under the heavy grazed treatment. Simulation of soil water at each soil depth and watershed was conducted using the Versatile Soil Moisture Budget Model (VB2000). Statistical and graphical evaluations of the model results were conducted using the volumetric soil water data collected for 1998 and 1999. The statistics determined included average error (AE), root mean square (RMS), coefficient of residual mass (CRM), modeling efficiency (EF) and coefficient of determination (CD). All statistics varied with each soil depth and watershed, indicating the transient nature of the data. This is reflected in the mostly negative CRM values, which ranged between −1.0 and 0.16. Overall model fitting to the whole data for all depths, watersheds and years gave values of CRM = −0.08 and EF = 0.19, indicating a slight over-prediction by the model. Spatial variation due to presence of rocks or cracks and averaging across slopes may have partly contributed to the discrepancies between model results and observed data.

Quantification and simulation of soil water on grazed fescue watersheds

A 2-year study was conducted at the Agriculture and Agri-Food Canada Stavely Range Substation, Alberta. The objective was to quantify and simulate the soil water status of small grassland watersheds under 3 grazing intensities and 4 topographic positions. The grazing treatments were ungrazed (or control), heavy (2.4 AUM ha-1) and very heavy (4.8 AUM ha-1) grazing and the topographic positions were upperslope, midslope, lowerslope and 5 m away from the collector drain. Moisture readings were taken every 2 weeks between spring and fall using a CPN 503 moisture neutron probe. Readings were taken at the soil surface and at 15-, 25-, 35-, 45- and 55-cm depths. Total annual precipitation in 1998 and 1999 was 648 and 399 mm, respectively. In both years grazing treatments did not affect total soil water in the 0-50 cm (TSW50) depth interval for the upper, middle and lower slope positions, but TSW50 close to the collector drain was significantly (P ≤ 0.05) greater for the heavy grazed compared to the very heavy grazed treatment. Within each grazing treatment, TSW50 differences among slope positions occurred mainly under the heavy grazed treatment. Simulation of soil water at each soil depth and watershed was conducted using the Versatile Soil Moisture Budget Model (VB2000). Statistical and graphical evaluations of the model results were conducted using the volumetric soil water data collected for 1998 and 1999. The statistics determined included average error (AE), root mean square (RMS), coefficient of residual mass (CRM), modeling efficiency (EF) and coefficient of determination (CD). All statistics varied with each soil depth and watershed, indicating the transient nature of the data. This is reflected in the mostly negative CRM values, which ranged between -1.0 and 0.16. Overall model fitting to the whole data for all depths, watersheds and years gave values of CRM = -0.08 and EF = 0.19, indicating a slight over-prediction by the model. Spatial variation due to presence of rocks or cracks and averaging across slopes may have partly contributed to the discrepancies between model results and observed data.

Patterns and simulation of soil water under different grazing management systems in central Alberta

A study was conducted at the Lacombe Research Centre to quantify and simulate the impacts of forage and grazing systems on soil water content. Four forages used in the study were alfalfa (Medicago sativa L.), a mixture of meadow bromegrass (Bromus riparius L.) and alfalfa, an annual pasture and an old grass pasture that was composed of mainly quackgrass (Elytrigia repens L.), smooth bromegrass (Bromus inermis L.) and Kentucky bluegrass (Poa pratensis L .). Within each 1.2-ha paddock were two grazing treatments: rotational grazed and ungrazed. Soil water measurements to a 65-cm depth were conducted between May and October of 1999 and 2000 using a neutron moisture probe. Total soil water was affected by forage species more than grazing. Actual evapotranspiration rates were 3-4 mm d-1 in both years. Simulation of daily volumetric soil water content (%) for each year was conducted using the Versatile Soil Moisture Budget (VB2000) model on grazed alfalfa, ungrazed alfalfa, grazed annual and ungrazed annual treatments. During calibration year of 1999, the overall modeling efficiency (EF) was 0.58 while, during the evaluation year it was 0.43. Further, simulations for alfalfa were better than those for annual treatments. These EF values are relatively low indicating substantial discrepancies between observed and simulated results, which could have been attributed to a combination of input data errors, model errors and propagation errors in output. Key words: Evapotranspiration, forages, model calibration, model evaluation, versatile soil moisture budget model

Impact of climate change scenarios on the agroclimate of the Canadian prairies

Regional climate change scenarios for the Canadian prairies were generated using historic weather data and daily data from two Canadian Climate Centre general circulation models (GCM). Model CGCM1-A was a recent version release while its predecessor was model GCMII. The GCM data were combined with historic values to generate two additional scenarios. All scenarios were used to drive the modified Versatile Soil Moisture Budget model that assessed soil moisture, aridity and other agroclimatic indices. The modelled results for all scenarios were compared to those using the historic climate data. The model predicted earlier seeding dates for spring wheat between 18 and 26 d. Early seeding and harvest was shown to be an appropriate adaptive strategy that avoided more arid conditions in the late summer. The soil water deficit was lower under GCMII than historic values by 46 mm. For CGCM1-A, the soil water deficit was decreased by 8 mm across the Prairie Provinces compared to historic values. GCM scenarios predicted unchanged or increased soil water in the top 120 cm soil across the Canadian prairies compared to the historic scenario. There were some regions such as south eastern Saskatchewan and southern Manitoba where reductions in summer rainfall (for CGCM1-A) were large. These regions experienced the greatest benefit of earlier seeding dates. Key words: Climate change, agriculture, aridity, growing degree-days, soil moisture, seeding date, harvest date

A climatological study of soil moisture under corn crop at Campina Grande (NE Brazil)

ABSTRACT. Results of a climatological study of soil moisture under corn crop at Campina Grande (NE Brazil) are presented in this paper. Daily values of available moisture content during the crop growing period are evaluated for a period of 25 years. A six zone versatile soil moisture budget model is used for this purpose and approximately 5. 7.5, 12.5, 25, 25 and 25% of the available water capacity (AWC) are attributed to zones one to six respectively. Different root activity coefficients are assumed for the six zones in different growth stages and the dependence of these coefficients on moisture content is taken into consideration. The same moisture releasing characteristics are assumed for all soil zones. On rainy days moisture loss due to evapotranspiration is assumed to take place before precipitation. Four AWC values and three corn growing periods between March and September are considered in this study. A first order Markov chain model is applied to the daily soil moisture data. Soil moisture averages and probabilities are used to identify the optimum growing period for corn at this station. The irrigation requirements of the crop are briefly discussed.

Observations and modelling of soil water content changes around trees

This paper presents the results of a 5-year study of water content variations in a clay soil due to tree roots and evapotranspiration. The site of the study program is Maisonneuve Park which is located in the eastern area of Montreal, in the heart of a zone known for settlement problems experienced by lightweight buildings supported on shallow foundations. The clay deposit around several trees (both isolated and in a row) has been monitored by means of several arrays of 3–5 m long thin aluminium tubes, spaced from 0.1H to about 3H from the trees, where H represents the height of the trees. The volumetric water contents around the tubes have been measured by means of a neutron probe. The study shows that while for some trees, the effects of the roots are confined to a volume of soil located at a distance less than the height of the tree, in other cases, the zone affected by the trees is of much greater extent, reaching as far as 1.5H. In addition, for some of the trees studied, their effect extends to depths exceeding 4 m. Water content changes have also been estimated using the Versatile Soil Moisture Budget model, which was developed at Agriculture Canada. Comparisons between soil moisture readings in the field with estimates from the model show the feasibility of estimating soil moisture changes from standard meteorological data. Key words: water content changes, clay deposit, tree roots, evapotranspiration, observations, modelling.

ESTIMATED SNOWMELT INFILTRATION AND RUNOFF FOR THE PRAIRIE PROVINCES

A revised version of the Versatile Soil Moisture Budget model, which included an algorithm for infiltration into frozen soils, was utilized to simulate snowmelt infiltration and runoff for the agricultural region of the Prairie Provinces of Canada. Thirty years of daily data from 78 climatological stations within the Prairie region were used in the model to produce maps of the average annual winter precipitation, snow water equivalent, and runoff for fallow and stubble lands.The average annual water equivalent of snowcovers within the region ranged between 25 mm and 60 mm on fallow land and between 35 mm and 85 mm on stubble land. Snowmelt infiltration did not vary widely throughout the region ranging from 17 to 23 mm for fallow and from 23 to 33 mm for stubble. Average annual snowmelt runoff ranged between 10 mm and 40 mm for fallow and between 15 and 55 mm for stubble. Average annual runoff for the three years with highest runoff was approximately double that of the 30-year average.

Relation between the normalized difference vegetation index and ecological variables

The objectives of this study were to analyze the relationship between the AVHRR data [transformed into a normalized difference vegetation index (NDVI)] and ecological variables and, secondly, to evaluate the possibility of estimating actual evapotranspiration using AVHRR and other readily available data. The study was undertaken across a range of conditions in Canada. Meteorological and composited NDVI data were compiled for 15-day periods over the 1986 growing season and were supplemented by digital ecological and soil data. The effect of soil and meteorological conditions on vegetation growth was modeled using the Versatile Soil Moisture Budget model, which accounts for additions and losses of water from various layers in the root zone. The various data sets were integrated using a geographic information system. We found that individual vegetation/soil combinations exhibited different NDVI trajectories, most closely related to the trends in potential evapotranspiration. There was a highly significant association between coincident NDVI and actual evapotranspiration values ( r = 0.77) or between NDVI and potential evapotranspiration for the previous 15-day period ( r = 0.86). High correlation was found between cumulative NDVI and cumulative actual evapotranspiration over the growing season ( r = 0.96). Actual evapotranspiration for individual 15-day periods could be estimated from NDVI and potential evapotranspiration values with a residual error of about 10–15%. The results show that the NDVI trajectories provide important information on the seasonal development of vegetation at northern latitudes. Results in estimating actual evapotranspiration were also encouraging but further work is needed to determine the accuracy limits for this geographic area. The representativeness of the 1986 results for the region is subject of continuing study with 6 years of NDVI data.