Agricultural Research Management Information System

Home
Research Summary
Search
About Us
- ARMIS
- Brochure
Contact Us
Report
User Request
Data Input
Help
- Operation Manual
  - PDF
  - Video
- Program Area & Commodity
We have reached 37600 number of research entries at this moment.

- Logout

Research Detail

Home
Research
Detail

A Multiplb Regression Model for the Prediction of Dry Season Moisture Contents of the Coastal Saline Rice Soils of Bangladesh

M K Mondal
Senior Agricultural Engineer
Irrigation and Water Management Division, Bangladesh Rice Research Institute(BRRI), Gazipur 1701, Bangladesh.

S I Bhuiyan
Former Water Resources Specialist
Soil and Water Sciencbs Division, International Rice Research Institute(IRRD, Los Banos, Philippines.

Abstract/ Executive Summary:

Experiments were conducted at Barodanga and Mirzapur villages under Dumuria upazila of Khulna district to document dynamics of soil moisture in the dry season of the coastal rice soils and develop a model for its prediction. The soil texture was silty clay loam to silty loam and soil moisture contents of the coastal saline rice soils in the dry season remained close to field capacity level most of the time except in the top layer. Multiple linear and non-linear regression models were developed to predict soil moisture contents in the dry season by using rainfall, evaporation and groundwater depth as independent variables. Soil moisture prediction was improved when non-linear model was employed in place of its linear form. Therefore, the nonlinear model may be used to predict soil moisture contents of the coastal rice soils and may be used for planning non-rice crop cultivation to improve land productivity of the coastal regions of Bangladesh.

Keywords: Multiple regression model, Prediction of dry season moisture content, Coastal saline rice soils of Bangladesh

Location: Barodanga and Mirzapur villages under Dumuria upazila of Khulna district

Start Date:

End Date:

Program Area: Knowledge Management

Commodity: Water management

Objectives:

To facilitate dry season crop production processes in the coastal regions of Bangladesh.

Methodology:

Southwest region of Bangladesh is the most problematic region as far as soil and water salinity are concemed. Most rice lands in this area are classified as moderately saline (EC_e - 4-8dS m^-1) and saline (EC_e = 3-16 dSm^-l). The villages Barodanga and Mirzapur, both in Dumuria upazlla of Khulna district were selected to represent the saline and moderately saline soil categories. Both sites are located at an altitude of 4 m above mean sea level. Evaporation, rainfall and groundwater level mainly govern the dynamics of soil moisture in coastal rice soils. Groundwater remains close to the surface in the dry season contributing to soil moisture through capillary rise. To predict soil moisture pattern of the coastal rice lands. Experiments were conducted at Barodanga and Mirzapur villages to monitor dynamics of soil moisture in the dry season in coastal rice soils. Five plots of 8m x 5m were selected for soil sampling at each site. Soil moisture contents of the experimental fields were measured at 15 day intervals. Bulk density and textural class of the experimental soil were also determined. Soil samples were collected in the middle of each plot from 0-15, 15-30 and 3045 cm soil layers to determine moisture contents and from 0-15, 15-30, 30-45, 45-60,60-80 and 80-100 cm to determine soil texture. For bulk density, soil samples were taken from one side of the experimental field. Undisturbed samples were taken from 0-15, 15-30 and 30-45 cm by a core sampler. Bulk density was determined by dividing oven dried sample weights by bulk volume of the soil sample, while soil texture and soil moisture were determined by hydrometer and gravimetric methods, respectively. Available soil moisture in the dry season was calculated from soil moisture and bulk density. Two observation wells were installed diagonally at two corners of the experimental field at each site to monitor groundwater depth at weekly interval. The average of these readings over a week was taken as the depth to groundwater for that week. A Taylor rain gauge (305 mm Clear-VU Rain and Sprinkler gauge with a rainfall capacity of 127 mm) and USWB Class A evaporation pan were installed in the vicinity of the experimental field for measurement of daily rainfall and evaporation rates, respectively. These devices were read daily at 9:00 AM.

Source of Information: Journal of the Institution of Engineers, Bangladesh, Agricultural Engineering Division, VoI.31/AE, No.1, Dec, 2004

Article Link (Online Journal):

Funding Source:

Total Budget (Tk.) :

Achievement/Findings:

Multiple linear and non-linear regression models were developed to predict dry season moisture contents (SMC) of fallow lands in coastal rice soils. The independent variables of the models were total rainfall (TRF), total evaporation (TEV), and water table depth (WTD). The nonlinear model provided better soil moisture prediction than the linear one. As the soil moisture is the key factor for crop cultivation in the dry season, the model could provide planning guidelines to timely establishment of dry season crops in the fragile coastal environments and could also provide information on irrigation and drainage needs of non-rice crops for optimum yields.

Knowledge Management: Journal