Regional Water and Soil Assessment for Mapping Sustainable Agriculture
	Sub-Project 1: Water Balance Modelling

• Background and Objectives
  
• Specific Objectives
  
• Year 1 (97/98) Progress
  
• Year 2 (98/99) Progress
  
• Year 3 (99/00) Progress
  
• Year 4 (00/01) Progress

Background and Objectives

The water balance Sub-Project will establish and validate an operational water balance model using data collected at the main sites at Luancheng and Changwu in China and the Liverpool Plains, MLR and DT sites in Australia. It will include new components to better model crop yield and be used to develop water saving, recharge and waterlogging scenarios with the calibrated models. This will form a basis for the development of water utilisation indicators.

The first step will be to review related research in the Chinese and English literature on water-saving agriculture and regional water balance appropriate to the sites in both countries. It is important to capitalise on the attention that has been given this area in the two countries and learn from the different responses. This will be followed by collating the field data for water balance models (climate, soil hydraulic data, vegetation) for the calibration sites. The meteorological, soil hydraulic, and crop biophysical data will be assembled for calibrating and validating the water balance model(s).

The next step will be to modify the plant growth component of the WAVES model so that it can predict crop yield and develop field measurable indicators of crop water stress. In particular, we will apply or implement environmental models of the physical, chemical, and biological processes in the root zone that are sensitive to the effects of management. The chosen methods need to effectively model the dynamics of water movement and solute transport in soil-plant-atmosphere continuum in the target environments. Once this is in place, scenario modelling will be used to examine the effect of cropping practice, water saving strategies and (where relevant) irrigation management on productivity at the site scale. The study of extension to less well instrumented sites (such as weather stations) and the issue of up-scaling will be developed in conjunction with the Information Systems Sub-Project. However, it is likely this will entail some development of appropriate simplifications to the models.

It is essential from the earliest stage to develop interactions with local authorities and agricultural managers to obtain information on the current cropping and (where appropriate) irrigation practices. The next task will be to calibrate the water balance model in terms of soil moisture storage, crop water use and growth using data obtained from calibration field stations at Luancheng, Nanpi, Changwu and Ansai in the NCP and LP areas of China and the LPL, MLR and DT areas in Australia. This will provide better understanding of the relationship between soil moisture availability and crop growth and provide confidence for use of the model over the associated regions. In particular, locally calibrated benchmarks for crop performances will be established as local WUE indicators.
 
 

Specific Objectives

Establish calibrated and validated water balance models for assessing levels of water use efficiency in dryland and irrigated systems.

Establish water partitioning between runoff, evaporation, transpiration and drainage for the landscapes being studied.

Establish models with improved ability to assess the relationships between soil water availability, soil properties, crop water use and crop yield in the agricultural systems of the study.

Evaluate scalable indicators of WUE (such as the locally calibrated WUEI) taking into account the effects of different crops, agricultural practices and climate variability on water balance and water conservation.

 
Year 1 (97/98) Progress - as extracted from the Annual Report

Literature review on water balance modeling approaches:

a. WAVES Documentation (CSIRO)

WAVES is a process-based model that simulates energy, water, carbon, and solute balances of a one-dimensional soil-vegetation-atmosphere system. The model was developed in Sustainable Catchment Management Program CSIRO Land and Water. As part of the ACIAR project, we have documented the WAVES model to provide detailed information on the concept and process representation. The document also includes program structure, numerical solutions, model testing, and clear worked examples. It summarized our research efforts and findings in the area of energy and water balance modelling. This document will greatly help users to understand the model and apply it appropriately for solving water balance problems.

Zhang, L. and Dawes, W.R. (1998). WAVES — An integrated energy and water balance model. CSIRO Land and Water Technical Report No. 31/98, pp. 177.

b. Study of the soil hydraulic properties in the North China Plains (Canberra and SIAM)

Soil hydraulic properties are important parameters for water balance modelling. A number of soil samples were taken from the North China Plains and analysed for their hydraulic properties. Based on these data, we compared three soil moisture retention functions and the relationship between soil moisture content and hydraulic conductivity. A paper was drafted in Chinese describing the results.

Zhang, X.Y. and Zhang, L. (1998). Estimating soil water retention and hydraulic conductivity using parametric functions — Examples from the North China Plain (in preparation).

c. Water balance modeling (IWSC)

A literature review on water balance modelling was carried out. A brief review of field water cycling and modeling approaches written in Chinese in the first year of the ACIAR project (Huang Mingbin, 1998, book chapter). Soil hydraulic properties are required for water balance modeling. An integral method was used for estimating soil hydraulic properties. A paper titled "Integral method for estimating soil hydraulic properties" was published in SSSAJ (Shao et al., 1998, 62: 585-592). This method is applicable to field conditions.

d. Water balance modelling (SIAM)

The application of the WAVES model at Luancheng Station has been carried on since the beginning of the project with the data input from some past and present projects. And some field soil moisture measurements are still undergoing with the support of other projects. Till now, the simulation has suggested that the WAVES model can efficiently simulate the crop transpiration and soil evaporation for winter wheat and corn, as well as the dynamics of soil water contents based on ten years results at Luancheng Station. This provided the basis of applying the WAVES to field irrigation management in NCP. Next step, efforts will be concentrated on deciding of irrigation index for different crop such as the critical soil water level and the sensitivity of crops to water stress at their different growth stages in combination with the WAVES for efficient farmland water management.

e. Study of water use efficiency

The precise meaning and methods to measure water use efficiency has varied between sites. Reuter et al has produced a discussion paper on the topic.

Based on the study of field water balance at IWSC research sites, an alternative irrigation method was developed to increase water use efficiency of corn crops. A manuscript by Kang et al was published in Agricultural Water Management. The irrigation method has the potential to improve water use efficiency in irrigation areas of the Loess Plateau.

f. Solute transport in soil

Boundary layer theory was used to solve convection-dispersion equation (CDE). The solution has potential to estimate solute transport parameters in field condition. A paper titled "An approximate solution to the convection-dispersion equation of solute transport in soil" was published in Soil Science (Shao et al., 1998, 163:339-345).

g. Data collection/availability & quality control

Metadata sets have been compiled for each research site in Australia and China. Assessments have been made about the suitability of the data sets for running WAVES. In China some data collection has been undertaken and these data are suitable for this sub-task and will be made available to the project. For example at SIAM the following was completed

• Ten year observation data such as meteorological data, hydrological information, soil water and nutrient dynamics, and etc. (1986-1996) from Luanchench Station and Nanpi Station were collated.
• More than 10 maps with different information from Nanpi and Luanchench Counties were digitized. The agricultural and economical statistics were collated.
• A lot of information, which is need for future study of GIS in NCP, were downloaded and verified from NCDC.
• Basis on the old soil classification map of NCP, a new classification standard (American Standard) was used to change the old classification system (former Soviet Union Standard) to meet the need of this project. Also, groundwater map have been revised to meet our unity purpose.
• Cooperated with Institute of Remote Sensing, CAS, some procedures were developed to assess the production of wheat and maize in NCP.

h. Include crop yield and stress in WAVES

A major effort has been put into this area and a module built into the WAVES model. Application of the APSIM modelling approach has been considered and an effort will be made this year to investigate synergies. The main problem with APSIM is that is does not have the capability to solve the energy balance which is required to link the water balance model with thermal remote sensing. This is essential to up-scale the plot-scale water balance estimates to provide regional measures.

• Canberra and SIAM : To study the effect of irrigation on water use efficiency, we have incorporated crop yield prediction into WAVES. Firstly, crop yield is simulated based on its relationship with above ground dry matter or biomass and this method requires only one parameter. Secondly, crop yield is calculated based on total water use during the period of growth. These two methods were chosen because they have the same level of complexity as the other processes represented in WAVES and require minimum changes to the program structure. Lu Zhang and Xi Ying Zhang reviewed the literature on crop yield prediction in relation to water use efficiency.

Wang, H.X., Zhang, L. Dawes, W.R., Liu, C.M. (1997). Improving water use efficiency of irrigated crops in the North China Plain — Measurements and modelling. Presented at the Soil Erosion and Dryland Farming Conference, Xi’an China, 15-19, September 1997.

• At ISWC : A simplified model of crop production was developed. The code written in CSMP (Continuous Systems Modeling Program) for solving the model is also provided for crop production prediction. This work will be related to water use efficiency indicator.

Year 2 (98/99) Progress - as extracted from the Annual Report

Following last year’s visit by Lu Zhang, the Yangling Group has managed to test the WAVES model using their data obtained from Changwu. A visit by Dr Shao Mingan to Canberra planned for November this year will summarise the results of this study. There are 13 papers published from the water balance sub-project. These papers deal with a range of issues related to water balance and water use efficiency. The Yangling Group looked at the effects of soil moisture and CO₂ concentration on evapotranspiration for some major crops in the Loess Plateau. They also studied the impact of irrigation schemes on crop yield. One paper from the group developed a new analytical solution for heat conduction-convection equation.

The Shijiazhuang Group focused its research on water use efficiency, the concept and theoretical background was reviewed. These papers also reported some recent developments in water use efficiency. Currently, most of research on water use efficiency for the North Chin Plan has been conduced at point or plot scale. It is important that we understand the key processes affecting water use efficiency at these scales. However, what is more important is to be able to provide information at regional scale, e.g. the entire North China Plain, for management purposes. This requires linking water balance and remote measurement process understanding. The development of crop yield and water stress in WAVES has been delayed as a result an internal CLW Divisional Review of water balance models. Currently methods for predicting crop yield have been incorporated into WAVES, these will be calibrated and validated soon.

The Adelaide Group studied the water balance of three representative texture-contrasting soils in the Keyneton Catchment in the Mt Lofty Ranges over the long-term (25 year) using a one-dimensional model. The results of the modelling showed that graziers can reduce deep drainage by as much as 50% by replacing the current cocksfoot, annual-based pastures with perennial based pastures (phalaris or lucerne) in all positions in the landscape. A 50% decrease in deep drainage in the Keyneton catchment, from the introduction of rotationally grazed perennial pastures, could reduce the long-term (25 year) average annual deep drainage from an estimated 50 mm below existing pastures to 25 mm. Thus, the introduction of rotational grazed perennial pastures must be coupled with other higher water using strategies such as tree plantations and engineering strategies to lower saline groundwaters. Studies on intact soil cores taken from the Keyneton catchment showed that deep drainage occurred in some of the soils through a well-developed macropore system. Results of two-dimensional modelling coupled with some measurement of flow within the Keyneton catchment supported the key findings of the 1-D modelling as well as showing the significance of throughflow.

The Canberra Group studied the effect of vegetation changes on catchment water balance. Many environmental problems in Australia are caused by changes in catchment water balance as a result of clearing native vegetation. It is now well established that tree plantations will increase catchment evapotranspiration compared with pasture and crops. This has implications for catchment water balance in terms of land use management and rehabilitation strategies to reduce recharge and salinisation. It is highly desirable to develop a method to predict the effect of such changes on catchment water balance. Based on previous studies and data from over 250 catchments around the world, a simple water balance model was developed. The model showed satisfactory results compared with filed data. Hence, the method is a practical tool for predicting the effect of vegetation changes on catchment water balance.

Year 3 (99/00) Progress - as extracted from the Annual Report

Strong links have been developed by scientists involved in water balance modelling in China and Australia. Following her visit to Canberra two years ago, Zhang Xingying published a journal paper on the use of different models for representing soil hydraulic properties. In this paper, she tested three commonly used models using data from the piedmont of Mt. Taihang. The results from this paper will help future water balance studies to better understand pant-water relationships for the North China Plain. See both SIAM_03 and SIAM_04 in the reference list.

The paper by Wang et al examined relationships between irrigation, ET, crop growth, and water use efficiency of a corn-wheat rotation in the NCP. It showed that different irrigation schemes are required for winter wheat and corn mainly due to rainfall distribution. This study found that soil evaporation is a significant proportion of total water use and can be reduced by mulching. The output from this work is listed in SIAM_01, SIAM_02, and CBR_01 in the reference list. Scientists on the North China Plain has also started to analyse groundwater levels (SIAM_05).

Dr Shao Mingan visited Canberra in January 2000 and worked with Lu Zhang on the test and application the WAVES model to the Loess Plateau. A journal paper has been submitted and it summarises the results of this study. It showed that the WAVES model was able to predict winter wheat growth and soil water balance. It also showed that the model can be used to provide information for agricultural water management in the Loess Plateau. Refer to CBR_02 in the reference list.

A visit to Canberra by Kang Shaozhong in April 2000 resulted in a journal paper. In this paper, we examined relationships between crop yield, irrigation, and water use efficiency. The results showed that evapotranpsiration, crop yield, and water use efficiency depend on the status of soil water content in different growing periods. Relationships have been developed which will allow prediction of crop yield and water use efficiency using the WAVES model. The study also recommended that a double-mile soil drying in the early vegetative growth period and severe soil drying the maturity stage of winter wheat is the optimum limited irrigation regime in this region. See CBR_03 in the reference list.

In CSIRO Land and Water Adelaide, working in the Mt Lofty Ranges, South Australia, there has been a great deal of output from scientists involved in water balance modelling. These following two paragraphs relate to references ADL_01 to ADL_03.

Rainfall was average to below average during the years of field trials in the Keynes catchment in the Mt Lofty Ranges, South Australia. Overland flow was generally only a small component of the water balance (less than 1% of annual rainfall but as high as 12 to 14% in two plots in the wettest year). Throughflow was generally higher than overland flow (less than 5% of annual rainfall but as high as 13% in the wettest year). Generally there was a trend for throughflow to increase downslope. Throughflow water had far higher concentrations of agricultural pollutants than overland flow and thus cannot be ignored as a major pathway for pollutants in streams in these environments. Deep drainage was as high as 20% of annual rainfall on the mid slopes under the unimproved pastures that farmers have sown. In contrast improved pastures were shown to minimise deep drainage and use deep stored soil water (from below 2 m) (ADL_02, ADL_03).

Recharge in the Keynes catchment caused groundwater levels to fluctuate by an average of about 1.5 m each year. Even in summer, the groundwater within the valley area was within a few meters of the soil surface. Thus, each winter the valley soils become saturated from groundwater as they have only a thin unsaturated zone in which to store infiltrating winter rainfall. Thus the spatial extent of saturation within the valley (and up the hillslopes) is related to the amount of “excess” rainfall less precipitation (and the small amount required for soil storage) and the capacity of the erosion gullies to discharge the water. Land degradation is not due to rising groundwaters as reported in other parts of Australia but rather the seasonal spatial extent of soil saturation (ADL_01).

The water balance scientists in Yangling, working on the Loess Plateau have studied many aspects of crop growth management, including the timing and components of fertiliser applied, the effect of mulch, and timing of irrigation amounts. Refer to paper ISWC_09 to ISWC_17 in the reference list. The approaches used range from data analysis of agronomic results to mechanistic modelling.

Year 4 (00/01) Progress - as extracted from the Annual Report