Last update: November 24, 2010 09:35:34 AM E-mail Print

 

Using climatic variables to model Nama Karoo subshrub dry matter production 


PCV du Toit

National Department of Agriculture

Grootfontein Agricultural Development Institute, Private Bag X 529, MIDDELBURG 5900

 


Introduction 

Models aid us in understanding and predicting the behaviour of complex systems. A relatively simple model is proposed to predict the complex dry matter production responses of Nama Karoo subshrubs in response to mean climatic variables. Previously published crop production models are purposely ignored because these models use production growth functions obtained during different phenological stages of the crop plant’s growth cycle. These data do not exist for Karoo subshrubs. The published crop growth models also incorporate various soil variables, which is not known for the Karoo subshrubs. Karoo subshrubs grow over a wide area on differing soil types. In the currently proposed model, soil, whatever its geological origin, is simply regarded as a growth medium for the shrubs, in order to keep the model as simple, but yet as functional as possible.

Method 

Aboveground available dry matter production for ten Karoo subshrub species were measured on a monthly basis, from August 1966 to January 1970, at the Grootfontein Agricultural Development Institute. The following species, classified into their respective acceptability classes to small stock, were studied. Desirable, palatable subshrubs included; Felicia muricata, Phymaspermum parvifolium, Plinthus karooicus and Salsola calluna, while less desirable, less palatable subshrubs were; Eriocephalus ericoides, Eriocephalus spinescens, Pentzia incana and Pteronia glauca and, undesirable, unpalatable subshrubs were; Chrysocoma ciliata and Eberlanzia ferox. Nomenclature follows Arnold and De Wet (1993).

A cocoon of bird wire-netting was fashioned around the perimeter and over the top of the canopy of each of the subshrubs, minutely following the outline of the shrub so that the cocoon fitted snugly around the subshrub. All the material that subsequently grew outwards through the cocoon of wire-netting, was harvested on a monthly basis.

The monthly aboveground available dry matter production values were then correlated with climatic data recorded at the Grootfontein weather station during the same period (Fig. 1). The climatic variables included in this model are: rainfall, temperature, wind run, evaporation rate and sunshine hours.

During the winter months; May, June, July and August, it is reasoned that only 50% of the rainfall contributes to the predicted production on account of the windy conditions leading to very high evaporation rates and desiccation despite the low temperatures experienced. In autumn; March and April, and in spring; September and October, it is reasoned that 50% of the last month’s rainfall plus 50% of the previous month’s rainfall contribute to the predicted production. During spring and autumn relatively mild temperatures and windless conditions are experienced, these favourable conditions lead to improved dry matter production and efficient use of the available water. During summer, however; November, December, January and February, 50% of the last month’s rainfall plus 50% of the previous month’s rainfall plus 25% of the rainfall received two months earlier, contribute to the predicted production. The influence of favourable rainfall events and follow-up rainfall events over a three months’ period on total dry matter production is accounted for by including variable amounts of rainfall to predict the outcome of subshrub dry matter production in summer.

The calculated monthly rain values are multiplied by 3. This value is the production estimate per mm of rainfall received in arid savanna grassland areas (Le Houérou 1984)(R - refers to the rainfall value in the model).

Mean temperature is calculated as the maximum temperature, minus 0.25 times the maximum temperature minus the minimum temperature (T refers to mean temperature in the model).

A mean daily value is calculated for each month of sunlight hours (S), evaporation rate (E) and wind run (W) (in the model, S refers to mean sunlight hours, E refers to mean evaporation rate and W refers to the mean wind run).

Since during any month, only 50% of that specific month’s rainfall is used in the model, it is not unreasonable to correct for this action by multiplying the equation by two. This action also results in the predicted model and the actual, measured production curve having the same units of measurement (Fig. 1).

Results 

The following model is proposed to predict aboveground available dry matter, making use of the climatic variables.

The model predicting aboveground available dry matter production = {([S + T] x R)÷(W x E)} x 2.

The predicted dry matter production curve has a correlation coefficient of 0.8130 with the actual measured dry matter production curve (Fig. 1). The d-index of agreement between the two production curves is 0.7374.

Discussion 

A model should include some of the most important functional attributes of the real system, while it is clearly understood that not all these attributes will necessarily be accommodated in the model. The model should be simpler than the real system to be of value in describing the real system. Therefore, constraints suffered by some models are that they need too much data input of a complicated nature, which must necessarily be collected over long time periods. The PUTU suite of models was perceived to be too complicated to use in the present modelling exercise, since it inter alia, includes soil pH, clay content and water availability, vegetation basal cover and daily carbohydrate reserve status and daily values of rainfall, sunshine duration, minimum and maximum temperature, to simulate forage production in order to predict current stock numbers that can safely be run on the veld on a daily basis (Fouché 1992; Howard 1993; Kellner & Booysen 2000).

 

Fig. 1 The measured monthly aboveground available dry matter production curve for four seasons, compared to the production curve predicted by the model

Plant production dynamics in the Nama Karoo are driven by variations in rainfall. Rainfall in the arid and semi-arid areas is therefore perceived as the principal factor influencing dry matter production, however, the inter-relations between numerous environmental factors should not be lost sight of. Yield of available dry matter is not necessarily related to standing crop in the Nama Karoo, because of the long turnover time in aboveground phytomass experienced with Karoo subshrubs. In situations where yield is important, as it is in the calculation of grazing capacities and the consequent estimation of current stocking rates, it should be known how much new available phytomass can be produced rapidly. Grazing capacities, deduced from current aboveground available phytomass production, is positively related to rainfall. In the planning of resource utilization, it would therefore be valuable if veld managers could predict peak yields in dry matter production during the growing season, in order to calculate stocking rates to be applied during the current grazing season (Duncan & Woodmansee 1975; Fouché 1992; Krebbs 1985; Pumphrey 1980; van den Berg 1983.

Measuring aboveground available dry matter produced by Karoo subshrubs in shorter time steps than one month, are extremely difficult and time consuming. It is clear that the method of harvesting, under-recovers the monthly production, on account of the fact that the material growing inwardly cannot be harvested and remains unaccounted for. This fact was, however, ignored in the present model since it is assumed that very little material does indeed grow inwardly and that by including it, it would have little affect on the model.

For this modelling exercise, climatic variables were classified as having either a positive or a negative effect on total dry matter production. Positive variables were; rainfall, sun and temperature; it is reasoned that with sufficient rainfall in the presence of abundant sunlight and favourable temperatures, plant production will be optimal. On the negative side; high evaporative demand, puts the plants in a state of negative water balance which is exacerbated by wind. The stronger the wind the more pronounced the effect. The especially parching north-westerly winds desiccates the plants by rapidly removing transpired water and in so doing, slows down production. Positive variable values, estimating dry matter production were divided by the negative variable values, in order to estimate lower dry matter production where the negative variable values were large and, conversely, to estimate higher dry matter production where the negative variable values were smaller.

The reasoning with the current model is that low rainfall, despite favourable temperature and sunlight, leads to low dry matter production and, that when no rainfall is received it does not, however, lead to nil dry matter production. Although the plants may look dead, they are ticking over, producing minute amounts of dry matter and waiting for favourable conditions to grow out actively. High rainfall on the other hand does not necessarily lead to a high production of dry matter by the subshrubs studied, there seems to be a ceiling above which the subshrubs in the arid areas cannot respond to increased rainfall. This fact detracts to a certain extent from the ability of the model to more accurately predict dry matter production.

It is furthermore reasoned that during no month does the total amount of rainfall received during that particular month, contribute to the total dry matter production of the subshrubs during that month. This reasoning is on account of the lag-period experienced between rainfall received and the actual production of dry matter in response to that rainfall. The lag period can last for between 14 to 21 days in the False Upper Karoo (Acocks 1988). However, during the growing season (November to February), all the rainfall received is eventually accounted for in the model.

In the absence of rainfall during any particular month in summer, growth and therefore aboveground available dry matter production does not stop, on account of the carry-over effect that the rainfall received during the previous month or two months has on dry matter production. This carry-over effect can last for as long as 45 to 60 days in the False Upper Karoo (Acocks 1988). During this time Karoo subshrubs are actively growing, despite having had no rain for up to two months.

From logic, it follows that high mean temperatures coupled to high wind runs lead to elevated evaporation and transpiration rates and therefore, low dry matter production on account of the low amount of water available to the plant for these processes. Although these conditions prevail most often during July and August, desiccating north-westerly winds can occur during any time of the year, severely affecting aboveground available dry matter production by the Karoo subshrubs.

Conclusion 

Total available dry matter production by Nama Karoo subshrubs can be modelled successfully. These modelled dry matter values can easily be converted to current grazing capacities, by appropriate computation, where the number of plants per hectare is known.

References 

Acocks JPH 1988. Veld Types of South Africa. Memoirs of the Botanical Survey of South Africa no. 57. Government Printer, Pretoria.

Arnold TH & De Wet BC 1993. Plants of southern Africa : Names and distribution. Memoirs of the Botanical Survey of South Africa no. 62. Government Printer, Pretoria.

Duncan DA & Woodmansee RG 1975. Forecasting forage yield from precipitation in California’s annual rangeland. Journal of Range Management 28:327-329.

Fouché HJ 1992. Simulering van die produksiepotensiaal van veld en die kwantifisering van droogte in die Sentrale Oranje-Vrystaat. Ph.D. thesis, University of the Orange Free State, Bloemfontein.

Howard MD 1993. Simulation studies on Digitaria eriantha Steud. Subsp. eriantha at differing soil nitrogen levels. M.Sc. Thesis, University of the Orange Free State, Bloemfontein.

Kellner K & Booysen J 2000. Modelling populations and community dynamics in Karoo ecosystems: 224-230. In: The Karoo : Ecological patterns and processes. Editors: Dean WRJ & Milton SJ.

Le Houérou HN 1984. Rain use efficiency : a unifying concept in arid-land ecology. Journal of Arid Environments 7:213-247.

Krebbs CJ 1985. Ecology : The experimental analysis of distribution and abundance. Harper & Row, New York. Pp 800.

Pumphrey FV 1980. Precipitation, temperature, and herbage relationships for a pine woodland site in north-eastern Oregon. Journal of Range Management 33:307-310.

Van den Berg JA 1983. The relationship between long term rainfall and the grazing capacity of natural veld in the dry areas of South Africa. Proceedings of the Grassland Society of southern Africa 18:165-167.

 

Published

Karoo Agric Vol 4 (1)