The demand for timber products in South Africa, and consequently afforestation, is increasing. There exists, however, abundant experimental evidence that trees utilise more soil water than other dryland crops. Because water is limited in South Africa, decision makers therefore currently face the challenge of determining a socially, and economically acceptable afforestation management plan to enable the reconciliation of increased timber demand with scarce water supply. This challenge, and the subsequent decisions that need to be made, may be accomplished by making use of suitable simulation models to predict the impacts of the forest hydrological system on water resources. Currently, these impacts are assessed through an Afforestation Permit System (APS) which is based on a model now acknowledged to have become outdated. In this dissertation an enhanced ACRU Forest Decision Support System (FDSS), now called the ACRU Forest model, is developed and proposed as a tool for modelling forest hydrological impacts on water resources. Research for this study included a literature survey, fieldwork at two locations, viz. at forest irrigation trials at Mkuze in northern KwaZulu-Natal, and at forest site preparation trials near Ugie in the Eastern Cape, as well as the evaluation, for purposes of model development, of a series of workshops. Results from the fieldwork experiments show that large tree water use potentials are possible if water is not limiting, although a water supply threshold exists at about 1400mm.annum-1, above which diminishing growth returns occur. Furthermore, trees display improved growth on more intensive forest site preparations, but at the expense of higher water usage rates. A series of workshops which had as the main objective the extraction of expert knowledge by stimulating responses to prepared questions and by constructive discussion on relative issues pertaining to forest hydrological modelling, yielded valuable information. This information, together with that gleaned from the literature, the fieldwork and a new Quaternary catchment database for South Africa, was used to develop the ACRU Forest model. The PC-based ACRU Forest model has the potential to aid decision makers by providing an initial indication of the impacts of afforestation on water resources, within a matter of minutes. An example of the model's application is used to demonstrate its operation, relative accuracy and its potential benefits in simulating hydrological responses to afforestation. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 1995.

A computer model was developed in order to determine the effects of drainage practices on nutrient losses from level agricultural soils. The model performs a daily simulation of the vertical flow of water, nitrogen, phosphorus, and heat, and of the growth of crops. A water flow submodel calculates the depth of the water table based on daily predictions of evaporation, transpiration, flow to drains and ditches, and deep percolation. An original saturated-unsaturated flow algorithm is used to determine moisture infiltration, redistribution, and upward flow in the soil matrix, as well as bypassing flow in the soil macropores and horizontal flux between the soil matrix and the macropores, and surface runoff. Nutrient movement occurs by mass flow. Heat flow, nutrient biochemical transformations, and crop growth are determined by using well established relations. Field tests were carried out for a period of two years on an experimental site in the Lower Fraser Valley of British Columbia. The water table depth was measured on a continuous basis. Grab samples of drainwater and observation wells were obtained periodically and analyzed for nitrogen (N0₃-N, NH₄-N, and TKN) and phosphorus (P0₄-P and TP). The field results show a decrease in the concentration of all nutrients over the sampling period, and provide evidence that denitrification and bypassing flow are important mechanisms affecting the nutrient balance of this soil. These results were used to calibrate the model. An excellent fit of the observed water table profile and an adequate fit of the observed drain concentration of nitrate were obtained. The simulation revealed that bypassing flow is a very important transfer mechanism in this soil and must be included in order to obtain a satisfactory fit of the experimental data. A sensitivity analysis of the model showed that the patterns of moisture flow have a predominant influence on the rate of nutrient leaching. In particular, it was found that the nutrient concentration in drain water is a strong function of the hydraulic conductivity of the soil matrix and of the horizontal distance between the soil macropores, which control the ratio of moisture flow in the soil matrix to the macropore flow and the lateral diffusion of nutrients between the soil matrix and the macropores. The effects of four different drainage designs on nutrient losses were simulated over a period of two years for three different soils and two different nutrient distributions in the soil. It was found that there is a large difference between the amount of nutrients leached from drainage systems using different drainage coefficients. There was also a large difference in the response of two drainage designs based on the same drainage coefficient but using different depth and spacing of drains. Transient effects, as determined by the initial vertical distribution of the nutrients, were seen to remain dominant over the two year duration of the simulation. The model was found to be useful in explaining the apparent contradictions found in the literature assessing the effects of subsurface drainage on nutrient losses. The results from the model show these effects to be strongly site and condition specific. Furthermore, the model shows that soils and drainage designs that produce similar volumes of drain flow may exhibit very different leaching responses, and that drainage designs equivalent from a hydraulic standpoint can be very dissimilar in their potential for leaching nutrients. The model provides a tool which can be used to determine the appropriateness of different drainage designs in soils where minimizing nutrient losses is critical. / Science, Faculty of / Resources, Environment and Sustainability (IRES), Institute for / Graduate

Experimental studies on infiltration/soil-water movement processes are vital to better understanding movement of soil-water in the vadose zone. The objective of this experimental research was to investigate infiltration/soil-water movement processes utilizing laboratory experiments and computer modeling. Small scale laboratory soil box infiltration experiments were conducted and utilized for the improved parameterization of the Green-Ampt (GA) saturated moisture content parameter to produce an effective moisture content parameter (Be) for utilization in a modified GA model. By incorporating ⊖e values into GA modeling, modeling results showed greatly improved wetting front prediction across different soil conditions. A new soil packing method was proposed for replicating complex microtopographical surfaces with uniform bulk densities in laboratory soil box experiments which proved efficient and effective at accomplishing both objectives. A rainfall simulator and an instantaneous-profile laser scanner were used to simulate rainfall and quantify surface microtopography for experiments. The results clearly show the effect of microtopography on infiltration and soil-water movement characteristics. This offers valuable insight into infiltration/soil-water movement processes as affected by different soil and surface microtopographic conditions. / National Science Foundation (Grant No. EAR-0907588)

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Civil Engineering, 1980. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING. / Bibliography: leaves 150-155. / by Paul Christopher Damian Milly. / M.S.

Thesis. 1977. M.S.--Massachusetts Institute of Technology. Dept. of Meteorology. / Microfiche copy available in Archives and Science. / Bibliography : leaves 120-121. / M.S.

No description available.

No description available.

In the past, derivative-based optimization algorithms have not frequently been used to calibrate conceptual rainfall -riff (CRR) models, partially due to difficulties associated with obtaining the required derivatives. This research applies a recently- developed technique of analytically computing derivatives of a CRR model to a complex, widely -used CRR model. The resulting least squares response surface was found to contain numerous discontinuities in the surface and derivatives. However, the surface and its derivatives were found to be everywhere finite, permitting the use of derivative -based optimization algorithms. Finite difference numeric derivatives were computed and found to be virtually identical to analytic derivatives. A comparison was made between gradient (Newton- Raphsoz) and direct (pattern search) optimization algorithms. The pattern search algorithm was found to be more robust. The lower robustness of the Newton-Raphsoi algorithm was thought to be due to discontinuities and a rough texture of the response surface.

A risk analysis of drought and an assessment of irrigation requirements were ascertained for a wheat (Triticum aestivum L.) crop in Parana, Brazil, using 28 years of historical weather data. Two soil moisture models, The Versatile Soil Moisture Budget (VB4) and SWACROP models, were compared using data from six wheat cropping periods. The models showed good performance in predicting soil moisture contents, but SWACROP underpredicted soil evaporation and runoff, and VB4 did not separate evapotranspiration into its components. Therefore, a new soil moisture model was proposed. In the new model, a Darcy type equation was used to calculate fluxes in the soil profile, and inputs of daily rainfall and potential evapotranspiration were partitioned during the day using simple disaggregation methods. Crop growth input parameters, interacting with weather and soil inputs, were used to calculate a detailed output of the water balance components. The validation of the model showed predictions of soil water contents and evapotranspiration in close agreement with field data. / A crop yield model based on the stress day index approach was selected from an evaluation of seven crop-water production functions using wheat field data. This model was combined with the soil moisture model to assess risks of drought during the establishment and development of non-irrigated wheat crops with different planting dates. Irrigation management strategies were simulated to identify net system delivery capacities and application frequencies that promote maximum yield with minimum requirements of water. Yield reductions in non-irrigated wheat due to water stress varied between 16%, for early plantings, to 50%, for late plantings. Maximum yields with minimum applied water was obtained by the use of low intensity (5 to 10 mm) and frequent (3 to 5 days) irrigations. System delivery capacity requirements varied from 1.5 to 3.0 mm/day, according to planting dates.

Irrigation in 1980 accounted for approximately 52 per cent of the water consumed in Southern Africa. The need for planning water resources in the agricultural sector is therefore apparent. Much of Southern Africa's arable farming is carried out on land which, in terms of soil moisture availability to crops, can be described as marginal. Information on soil moisture is therefore valuable to the agriculturalist for planning irrigation schemes and for dryland farming. The objectives of this study were to provide the information mentioned above. This was achieved by producing a detailed delimitation of 712 zones throughout Southern Africa, of more or less homogeneous climate and by providing estimates of crop water requirements under dryland and irrigated conditions in each zone. At the same time the bulk of information which is normally forthcoming from such an analysis involving a large number of combinations of possible input, i.e. crops, soils and planting dates, was reduced, whilst the essential information content was retained. The study provided inter alia an estimate of the frequency of non-exceedance of certain levels of irrigation requirement, based on analyses of soil moisture budgets using long daily rainfall records. The soil moisture budgeting models which were used to estimate the above information were verified inter alia using field measurements of soil moisture. The irrigation analysis was designed such that the results should not became redundant when the inevitable improvement occurs in the estimation of crop factors or soil moisture variables nor if the farming practices change with respect to planting dates. A dryland soil moisture budget analysis for a range of crops and soils was performed in addition to the abovementioned irrigation analysis. The need for this latter study stemmed from the belief that irrigation should not be considered in isolation but rather as one of a range of options, many of them involving dryland farming, facing the agriculturalist. In addition to the dissertation, this study produced a map of Southern Africa on which the 712 homogeneous climate zones are depicted. For each of these zones four pages of computer printout were produced. These pages contain the results of the crop water requirements study for irrigated conditions and the crop water requirement deficit, runoff and an index of stress days for a range of crops, soils and planting dates, under dryland conditions. / Thesis (Ph.D.)-University of Natal, Pietermaritzburg, 1988.