Masters Degrees (Agrometeorology)

Permanent URI for this collectionhttps://hdl.handle.net/10413/6565

Browse

Now showing 1 - 20 of 26

Comparative evaporation measurements above commercial forestry and sugarcane canopies in the KwaZulu-Natal Midlands.
(1999) Burger, Caren.; Savage, Michael John.
An understanding of the water use of different crops commonly grown in an area is essential for the implementation of integrated catchment management in South Africa. With increasing pressure on water resources, mainly due to the recent changes in the Water Act, it has become important to determine the actual water demands of agricultural and other crops. Policy makers require knowledge of whether forestry canopies use more water than grassland and other agricultural crops. The Bowen ratio and Penman-Monteith methods were used in a comparative study of the evaporation from Saccharum, Acacia and Eucalyptus. All of the research was conducted at marginal sites located in the KwaZulu-Natal Midlands of South Africa over a period of two years. The Bowen ratio energy balance (BREB) technique combines the Bowen ratio (J3) (the ratio between the sensible, H and latent heat flux density, XE), with the net irradiance (Rn) and soil heat flux densities (G) to calculate evaporation. A comparative study of the sitespecific energy balance components (Rn, G, H and AE), general climatic conditions (rainfall, solar irradiance and air temperature) and other site-specific parameters (leaf area index and average canopy height) was conducted on Saccharum and young commercial forests consisting of Acacia and Eucalyptus. The energy balance highlighted important differences in the energy balance components between the different canopies. The differences between the reflection coefficients at the three sites contributed mainly to the differences in the evaporation rates. The low reflection coefficients of the forest canopies (Acacia and Eucalyptus) (0.1 and 0.08 respectively) were smaller than of the sugarcane canopy (0.2). This resulted in more energy available (« 6 %) for partitioning between the sensible and latent heat flux densities and higher evaporation rates for the forestry canopies. Where low leaf area indices existed (Acacia and Eucalyptus sites) (LAI < 2), the soil heat flux density contributed up to 40 % of the net irradiance (G = 0.4 Rn). The evaporation rates for Saccharum, Acacia and Eucalyptus averaged 2 mm day"1 in winter and 5 mm day"1 in summer. The slightly higher summer evaporation rate for Eucalyptus (5.6 mm day"1), compared to Acacia (4.9 mm day"1), resulted from the lower reflection coefficients and canopy resistance (rc) for Eucalyptus (ocs = 0.08, rc = 35 s m"1) compared to Acacia (ocs = 0.1, rc = 45 s m"1). Automatic weather station data (solar irradiance, air temperature, water vapour pressure and windspeed) were applied to site-specific Penman-Monteith equations to predict evaporation for all three sites. Statistically significant relationships (slope, m « 1, r2 > 0.8) were found between the measured (Bowen ratio) and simulated (site-specific Penman-Monteith) evaporation estimates. The current study has demonstrated the effectiveness of applying the Penman-Monteith equation to forest and sugarcane canopies to predict evaporation, provided accurate net irradiance, soil heat flux densities and canopy resistances are used.
Varying levels of incident solar irradiance and microclimatic variations on banana (Musa spp) growth and productivity.
(2001) Kizito, Fred.; Savage, Michael John.
A field experiment was conducted at Inselele, KwaZulu-Natal South Coast, South Africa, in 1999/2000, to assess the influence of shading as related to varying levels of incident solar irradiance as well as microclimatic variations on banana (Musa spp) growth, phenology trends, morphology and productivity. The trial was established in August 1999 on a ratoon plantation. The experimental site, 0.655 ha in extent, comprised of three replications with four treatments having varying levels of incident solar irradiance levels of 100 %, 70 %, 40 % and 20 % under a planting density of 1666 plants ha. The irradiance levels were derived from black shade cloth, erected 1 m above the banana canopy level. Banana plantations have vegetation that does not completely cover the underlying ground surface. For such a canopy, there are basically two distinct and interacting surface components, the overstorey/canopy and the understorey/soil. Independent investigations and measurements of the solar energy fluxes for each of these two components forms a vital step to comprehend the factors that control the overhead energy fluxes in the plantation. In this study, evaluation of flux components in the understorey of the canopy using plastic microlysimeters was conducted. Considering normal variations in field measurements, the agreement and consistency among the different measurements with previous findings was adequate. Microlysimeter measurements of daytime soil evaporation were generally less than 1 mm, with an average of 0.45 mm. The study did not include flux measurement at the two level approach (Bowen ratio and Eddy correlation methods) which could have been compared with the single level approach and microlysimeter results. Two methods of deriving/measurement of energy fluxes were used and the differences between them are discussed. A reassessment of the microlysimeter technique is suggested. For the understorey, the sum of sensible and latent heat fluxes derived from spreadsheet computation was equal to the available energy. Mean soil temperatures at a depth of 20 to 60 mm ranged between 13 and 16 degrees C. This study illustrated that energy flux measurement and interpretation in within the experimental area. Good agreement was found in the patterns of wind speed profile measurements, with the 1.5 m profile depicting a mean difference of 52 % compared to the 3 m profile between day of year (DOY) 230 and 248 inclusive. This suggested that wind speed attenuation was strongly correlated to increment in height within the plantation due to canopy roughness. Differential canopy temperatures (measured with infrared thermometry) were more sensitive to the vapour pressure deficit than to wind speed. The most dense shade affected banana productivity indicated by a bunch weight of 22.69 and 33.65 kg under the 20 % and 100% irradiance treatments respectively. The bunch mass reduction was 32 %. Flowering dates were delayed by 8 days, 13 days and 21 days with incident irradiances of 70 %, 40 % and 20 % of the unshaded control (lOO % irradiance), respectively. The phenological responses in this study appear to be a result ofa contribution or interaction of both seasonal responses and shading treatments and this is further evidenced by the high levels of correlation (98.4%) reported between these two variables. Shading resulted in diminished leaf emergence rates (LER), pseudostem circumference and pseudostem height. However, just before flowering, no significant differences were observed in the pseudostem circumferences. There was a progressive increment in pseudostem height for all the treatments, with the 20 % irradiance treatment depicting the least heights registered compared to the rest of the treatments. Evident seasonal differences were registered in the LER and emergence to harvest (EH) interval. Comparison of LER and mean air temperature trends revealed a similar curve pattern and depicted a highly positive correlation of90.4%. The highest LER of3.8 was recorded in the month of January at a peak mean monthly air temperature of24.3°C while the lowest LER of 1.2 was registered in July which had the lowest mean air temperature of 14.7 cc. The EH intervals measured between September and December flowering varied from 125 days to 112 days respectively before harvest yet the April to May flowering had an EH duration of186 to 195 days respectively. The strong seasonal influence on phenological responses is further confirmed by these EH trends. Reductions in LAI observed with time were principally due to leaf senescence.
Modeling variety differences in canopy growth and development of sugarcane (Saccharum officinarum L.) using Canegro.
(2003) Zhou, Marvellous Mabeza.; Savage, Michael John.
Crop models have great potential as research tools, for crop system management and policy analysis. One of the most promising future uses of crop models is in crop improvement. The limitation in the use of models for crop improvement has been the inability of crop models to predict variety differences. Currently, the CANEGRO model, a sugarcane crop model developed the South African Sugar Association Experiment Station (Inman-Bamber, 1991a) can only model the performance of the NC0376 variety. Experiments were undertaken in the South East Lowveld of Zimbabwe, which is a hot and dry environment where sugarcane is grown under irrigation, to examine the canopy growth and development of four commercial varieties, ZN6, ZN7, N14 and NC0376. The study aimed at determining variety differences in canopy (tillers and leaves) development, develop parameters that can be used to model variety differences and test the improved CANEGRO canopy model for its ability to predict variety differences in canopy growth and development. For the late season, the numbers of leaves and tillers produced by each variety were counted every fortnight throughout the crop cycle. The total leaf area of the varieties and the individual leaf area on a stalk were determined using a Delta-T leaf area meter every fortnight. The date of emergence of successive leaves on a stalk was recorded daily. The leaf angles of each variety were measured every fortnight. The amount of photosynthetically active radiation (PAR) intercepted by the varieties was measured using a SunScan Ceptometer. Tillering and tiller senescence rates, phyllochron intervals, extinction coefficients and base temperatures were determined for the growth and development processes of varieties ZN6, ZN7, N14 and NC0376. Tiller and leaf population development was varietal. Tillering and leaf emergence were highly correlated to thermal time while tiller and leaf senescence were less correlated to thermal time. The poor correlation of the senescence phases to thermal time could mean that tiller and leaf senescence was driven by other factors other than thermal time. PAR interception could be one of these factors. The data showed that PAR interception could be a trigger of tiller senescence. The study showed that the tiller and leaf population development could be approximated by two linear equations. Tillering will be the first linear phase and tiller senescence the second linear phase. The first linear phase is driven by thermal time. While the second linear phase is triggered by PAR interception, the major driving factors need to be determined. This study proposed the use of two linear equations to model tiller and leaf population development as opposed to the polynomial equations used in the current CANEGRO model. Polynomial equations assume the factors driving tillering and tiller senescence are the same. The green leaf numbers per plant showed that all varieties experienced a decline in green leaf numbers with crop age. Varieties NC0376 and ZN7 had the greatest decline in green leaf numbers per plant while varieties N14 and ZN6 had the least decline. Variety ZN7 had the highest number of green leaves per plant while NC0376 had the least. The tiller growth and development was divided into three phases: the exponential phase during the initiation of stalks, the first linear phase during a period of rapid stalk elongation and the second linear phase during sucrose accumulation and maturation. The first two phases of development were driven by thermal time while the sucrose accumulation was not. There were variety differences in tiller growth and development. There were variety differences in base temperature for the development of various components of the canopy. Internode formation occurred at lower air temperatures than stalk elongation and tillering while canopy heights were correlated with higher air temperatures. This implies that internode formation could occur under conditions unsuitable for stalk elongation and may explain the short internodes frequently observed in stalks exposed to winter during rapid stalk elongation. The basic requirements for physiological parameters are that they should be stable across different environments, have significant differences between varieties and have physiological meaning. The parameters studied were thermal time requirement for shoot emergence, leaf appearance, to reach peak tiller population and to start of stalk elongation; surface area of the youngest biggest leaf, leaf number of the youngest biggest leaf, PAR transmission at the start of tiller senescence, extinction coefficients, and peak and mature tiller population. The difference between varieties in thermal time to shoot emergence was least using a base temperature of 16 QC compared to using 10 QC and therefore 16 QC could be a more appropriate base temperature for shoot emergence. The accumulated soil temperatures were less variable than accumulated air temperature and could therefore be a more reliable driver of shoot emergence. However, the limitations in the use of soil temperature are that it is not a readily available measurement and that it is not easy to measure. The gradual increase in phyllochron intervals appeared to be a better method of predicting leaf appearance compared to using a broken stick model. The phyllochron gradient was proposed, as it is likely to be a more robust way of modelling leaf appearance. The varieties had different phyllochron gradients. Variety ZN7 had highest rate of leaf appearance and produced the highest number of leaves per stalk while NC0376 had the lowest rate and produced the least number of leaves. There were statistically significant differences between varieties (P = 0,05) in PAR transmission at the start of tiller senescence and a base temperature of 16 QC was best at determining accumulated thermal time to the start of tiller senescence. Varieties with higher peak tiller population had higher final tiller population, lower thermal time per tiller and a higher ratio of final to peak tiller population. There were differences between varieties in the youngest leaf number attaining maximum leaf area and the leaf area of the youngest biggest leaf. Variety Nl4 had the biggest leaves and NC0376 had the smallest. Variety Nl4 had the highest leaf area index (LAI) while ZN7 had the lowest. There were significant differences (P = 0,01) in PAR intercepted by the varieties but there were no significant differences in extinction coefficients. Extinction coefficients increased with crop age. The varieties had significantly different (P = 0,01) leaf angles and ellipsoidal leaf angle distribution parameters. The measurement of LAI using SunScan ceptometer provided a better estimate of extinction coefficients than LAI measured using Delta-T leaf area meter. Model evaluation showed that CANEGRO canopy model version 2 was improved compared to than version 1. The model (version 2) was accurate in predicting tiller heights and dead leaf numbers per stalk. It was fairly accurate in predicting green leaf numbers per plant, stalk population and intercepted PAR but was poor in predicting LA!. Version 2 has proved to be a substantial improvement over version 1 in predicting stalk population. Generally, the version 2 model overestimated tiller heights early and underestimated later, overestimated the tiller population and LAI after peak, underestimated green leaf numbers per stalk for varieties ZN6, ZN7 and N14 and overestimated dead leaf numbers per stalk and intercepted PAR. The version 2 model predicted a constant green leaf numbers per plant and LAI from peak to harvest while observed data showed that green leaf numbers per stalk and LAI decreased towards harvest. Version 2 model predicted the tiller population of NC0376 closely but underestimated tiller senescence in N14 and also underestimated final tiller population in varieties ZN6 and ZN7. Future model refinements may need to focus on the prediction of the sigmoid pattern of tiller heights. The model may need to be calibrated to predict the green leaf numbers per stalk accurately, which should possibly improve the prediction of LAI that in turn could improve the prediction of intercepted PAR. The improvement in the timing and rate of tiller senescence should improve the prediction of tiller population particularly in varieties ZN6, ZN7 and N14. The study showed that the broken stick method IS superior in explaining leaf and tiller population development compared to using polynomial equations. The development of variety parameters helped improve the prediction of variety differences in canopy growth and development. A major weakness of most crop models is modelling variety differences in canopy growth and development. The inability of crop models to predict variety differences has limited their use in plant breeding. This study has resulted in an improved version of CANEGRO version 1 that is an initial attempt at modelling variety differences of sugarcane.
Estimation of reference evaporation and comparison with ET-gage evaporimeter
(2003) Abezghi, Tekeste Weldegabrial.; Savage, Michael John.
Accurate estimation of reference evaporation is necessary for the estimation of actual evaporation for irrigation and water resource management purposes. Estimation of reference evaporati~n using the Penman-Monteith method using automatic weather station (AWS)measurements requires the available energy to be accurately estimated. The available energy of short grass of 0.12 m was measured using a component net radiometer and soil heat flux plate measurements at the Faculty of Sciences and Agricultural (Agrometeorological station, University of Natal, Pietermaritzburg, latitude ~29.79 oS, longitude ~ 30.95 °E, altitude ~ 650 m). In an attempt to evaluate the accuracy of commonly used procedures of estimating available energy, estimates of net irradiance (from net long wave irradiance and reflection coefficient estimate) and soil heat flux density were compared to the actual measurements. The linear approximation of atmosphere minus crop surface emittance based on air temperature was compared with measured net long wave irradiance and similar empirical formulations. The underestimation of the measured net long wave irradiance was observed using the linear approach. Furthermore, a plot of measured clear sky net long wave irradiance and air temperature showed a logarithmic relation. The estimated reflected solar irradiance was overestimated for the reference crop. The measured soil heat flux density was observed to vary not only with net irradiance but also with cloudiness, wind speed and soil water content. The soil heat flux density measured with plates was noticed to follow the measured net irradiance. The sensitivity of Penman-Monteith latent heat estimate was investigated for the use of estimated reflection coefficient and soil heat flux density as well as ignored soil heat flux density. Results showed the latent heat estimate to be greater when soil heat flux density was ignored. Reduced set assumptions of Penman-Monteith were assessed usmg the microclimatic measurements. The grass reference evaporation estimate using estimated water vapour pressure from the pervious day minimum air temperature and approximated wind speed were found to be seasonal and procedure dependent. The hourly-reduced set estimate of reference evaporation was in good agreement with the grass Penman-Monteith estimate. The estimated daily water vapour pressure underestimated the daily grass Penman-Monteith estimate. The sensitivity of the reduced set reference evaporation estimate was compared for the two values of approximated wind speeds. The assumption of 2 m S-1 wind speed gave a relatively better result. The sensitivity of the surface temperature energy balance (STEB) estimate of reference evaporation was investigated using two different atmospheric stability procedures. The evaporation estimate agreement and performance of the technique were found to vary depending on the stability correction procedure. The Monteith (1973) correction procedure was observed to be more sensitive to a higher surface-air temperature difference. The Monteith (1973) procedure was found to underestimate the reference evaporation and this resulted in a lower correlation coefficient. The uncorrected and Campbell and Norman (1998) stability corrected procedure of STEB estimate overestimated the reference evaporation but resulted in good agreement with actual reference evaporation. The use of estimated available energy using the STEB method resulted in a 7 % overestimate of measured available energy. Different designs of atmometers have been used to measure evaporation. The less expensive and simple ET_gageR (Model A and E) atmometer for daily evaporation measures were compared to grass-based and alfalfa-based Penman-Monteith and STEB estimate of reference evaporation. Two different evaporation surface covers used with the device allowed for the comparison to be made. Measurements using the canvas 30 ET-gage cover for grass reference evaporation were compared to grass based Penman-Monteith and STEB reference evaporation estimates. Correlation between the canvas 30 measures and Penman-Monteith estimates were good compared to the STEB estimate. The ET-gage canvas 54 measures were in a good agreement with alfalfa based Penman-Monteith reference evaporation estimate. There was, however, a slight time lag in ET-gage evaporation with ET-gage evaporation continuing accumulation when the reference evaporation was zero. The sensitivity of the ET-gage for microclimate variation was tested using the measurements made for two levels and three different microclimates. A shade measurement of reference evaporation was overestimated. The response of the ET-gage to one and two meter microclimate measures was similar to the short grass measurement. Furthermore, the ET-gage surface evaporation estimate using the STEB method showed equal response to the ET-gage surface for the microclimate measure and explained the possible cause of the lag of the ET-gage response. Accurate microclimate measurements is a requirement for the performance of the PenmanMonteith approach for the estimation of reference evaporation. The investment cost required for an AWS set up is high. Alternative options for gathering information of the microc1imate measurements required for calculating reference evaporation were assessed in terms of cost saving, accuracy and other advantages. A weather station system using a Hobo H8 logger (internal relative humidity and air temperature sensor and two external channels, one which was used for solar irradiance measurements) was found to be a cost-effective method for calculating the necessary microclimatic information for calculating reference evaporation. With this system reference evaporation was estimated with reasonable accuracy, at 16 % of the cost of normal AWS system. The use of an Event Hobo logger and an ET-gage was found to provide a reasonable estimate of reference evaporation. The use of the reduced set evaporation weather station was found to be unreasonable in terms of cost and accuracy. Air temperature and relative humidity were measured from different design of radiation shields and Stevenson screens. The use of home-made seven-plate plastic radiation shields provided a similar shield to radiation and ventilation compared to manufactured shields. At a low solar angle when wind speed was very low, all the radiation shields including the small Stevenson screens showed a higher air temperature difference relative to the standard Stevenson screen. The highest average difference of air temperature measurement was measured within the small Stevenson screen and metal-radiation shield. The home-made plastic radiation shield showed similar averages of air temperature and water vapour pressure difference compared to the six- and twelve-plate Gill radiation shields. The home-made metal radiation shield showed relatively higher deviation from the mean being cold at night time and hot during the day. More research is needed to explore the efficiency of the ET-gage evaporation from variety of microclimates to establish the cause of the overestimate under shade, to develop better relation of clear day net long wave irradiance and air temperature and the use of a wind speed sensor with Hobo H8 weather station system.
Estimating solar radiation for water-use and yield simulations under present and projected future climate using Cropsyst.
(2003) Abraha, Michael Ghebrekristos.; Savage, Michael John.
Agricultural scientists are faced with the challenge of producing enough food for the increasing world population. Hence the need to develop tools for managing soil and plant systems to increase food production in order to meet the world food demand in the future. Crop simulation models have become promising tools in predicting yield and related components fi'om a set of weather, soil, plant and management data inputs. This study describes the estimation of solar radiant density, a crucial input in crop simulation models; calibration and validation of a soil-plant growth simulator, CropSyst, for management purposes; and generation of weather data for assessment of crop production under possible climate changes in the future. Daily solar radiant density, an input required by most crop simulation models, is infiequently observed in many stations. This may prevent application of crop simulation models for specific locations. Long-term data records of daily minimum and maximum air temperatures, precipitation, sunshine hours and/or solar radiant density were obtained for Cedara, Durban, Seven Oaks and Ukulinga in KwaZulu-Natal, South Africa. Solar radiant density was estimated fi'om sunshine hours using the Angstrom equation and ten other models that involved daily minimum and maximum air temperatures and/or precipitation along with extratelTestrial radiant density. Coefficients for the Angstrom equation and one of the other ten models were specifically developed for South African conditions; the remaining models required fitting coefficients using the available data for all locations. The models were evaluated using (i) conventional statistics that involved, root mean square elTor (RMSE) along with its systematic and unsystematic components, slope, intercept, index of agreement (d), and coefficient of determination (R\ and (ii) a fuzzy expert system that involved a single modular indicator (Ira d) aggregated from the modules of accuracy (aggregation of the indices relative RMSE, model efficiency and I-student probability), con'elation (Pearson's correlation coefficient) and pattem (aggregation of pattem index vs day of year and pattem index vs minimum air temperature). For each index, two functions describing membership to the fuzzy subsets Favourable (F) and Unfavourable (V) were defined. The expelt system calculates the modules according to both the degree of membership and a set of decision rules. Solar radiant density estimated from sunshine hours for the Durban station resulted in R2 , RMSE (MJ m,2) and d index of 0.90, 2.32 and 0.97 respectively. In the absence of observed solar radiant density data, estimations from sunshine hours were used for derivation of coefficients as well as evaluation of the models. For Durban, the performance of the models was generally poor. For Cedara, Seven Oakes and Ukulinga two of the models resulted in a high d index and smallest systematic RMSE. The solar radiant density estimated from each model was also used as an input to simulate maize grain yields using the soil-plant growth simulator, CropSyst. The models were ranked according to their ability to simulate grain yields that match those obtained from using the observed solar radiant density. The rankings according to crop simulation, conventional statistics and expert system were compared. The CropSyst model was also evaluated for its ability to simulate crop water-use of fallow and cropped (oats, Italian ryegrass, rye and maize) plots at Cedara, KwaZulu-Natal, South Africa. Soil characteristics, initial soil water conditions, irrigation and weather data were inputted to CropSyst. Crop input parameters for oats, Italian ryegrass and rye were used, with little modifications, as determined from field experiments conducted at Kromdraai open cast mine, Mpumalanga province, South Africa. Crop input parameters for maIze were either determined fi'om field experiments or taken from CropSyst crop input parameters documentation and adjusted within a narrow specification range of values as dictated by CropSyst. The findings indicated that CropSyst was generally able to simulate reasonably well the water-use of fallow and cropped (oats, Italian ryegI°ass, rye and maize) plots; leaf area index and crop evapotranspiration of rye; and grain yield and developmental stages of maize. The validated CropSyst model was also used to simulate timing and amount of irrigation water, and investigate incipient water stress in oats, Italian ryegrass and rye. The CropSyst model was used to investigate potential effects of future climate changes on the productivity of maize grain yields at Cedara, KwaZulu-Natal, South Africa. The effect of planting date (local planting date, a fortnight earlier and a fortnight later) was also included in the study. A 30-year baseline weather data input series were generated by a stochastic weather generator, ClimGen, using 30 years of observed weather data (l971 to 2000). The generated weather data series was compared with the observed for its distributions of daily rainfall and wet and dry series, monthly total rainfall and its variances, daily and monthly mean and variance of precipitation, minimum and maximum air temperature, and solar radiant density. Four months of the year failed to reproduce distributions of wet and dry series, daily precipitation, and monthly variances of precipitation of the observed weather data series. In addition, Penman-Monteith reference evaporation (ETa) was calculated using the observed and generated data series. Cumulative probability function of ETa calculated using the generated weather data series followed the observed distribution well. Moreover, maize grain yields were simulated using the generated and observed weather data series with local, a fortnight earlier and a fortnight later planting dates. The mean simulated grain yields for the respective planting dates were not statistically different from each other; the grain yields simulated using the generated weather data had significantly smaller variance than the grain yields simulated using the observed weather data series. When the generated weather data series was used an input, the early planting date as compared to the locally practiced and late planting dates resulted in significantly greater simulated grain yields. The grain yields simulated using the observed weather data for the early and local planting dates were not statistically different from each other. The baseline period was modified by synthesized climate projections to create future climatic scenarios. The climate changes considered corresponded to doubling of [C02] from 350 to 700 ~t1 ,-I without air temperature and water regime changes, and doubling of [C02] accompanied by increases in mean air temperature and precipitation changes of 2 (lC and 10%, 2 (le and 20%>, 4 °c and 10%, and 4 (lC and 20% respectively. Solar radiant density was also estimated from daily air temperature range for all scenarios that involved change in mean air temperature. In addition, input crop parameters of radiation-use and biomass transpiration efficiencies were modified for maize, in CropSyst, to accommodate changes in elevated levels of [C02]. Equivalent doubling of [C02], without air temperature or water regime changes, resulted in increased simulated grain yields as compared to the baseline period. Adding 2 QC to the mean daily temperature and 10% to the daily precipitation of a [C02] elevated atmosphere reduced the grain yield but still kept it above the level of the baseline period grain yield. Adding 4 QC to the mean daily temperature and 10% to the daily precipitation fLllther decreased the yield. Increasing the daily precipitation by 20% instead of 10% did not change the simulated grain yield as compared to the 10% increments. Early planting date, for all scenarios, also resulted in higher yields, but the relative increment in grain yield was higher for the late planting dates with scenarios that involved increment in mean air temperature. In general, this study confi1l11ed that doubling of [C02] increases yield but the accompanied increase in mean air temperature reduces yield.
Frequency domain reflectometry for irrigation scheduling of cover crops.
(2003) Gebregiorgis, Mussie Fessehaye.; Savage, Michael John.
A well-managed irrigation scheduling system needs a rapid, preCIse, simple, costeffective and non-destructive soil water content sensor. The PRl profile probe and Diviner 2000 were used to determine the timing and amount of irrigation of three cover crops (Avena sativa L., Secale cereale L. and Lolium multiflonlm Lam.), which were planted at Cedara, KwaZulu-Natal. The PRl profile probe was first calibrated in the field and also compared with the Diviner 2000. For the calibration of the PRl profile probe the factory-supplied parameters (aJ = 8.4 and ao = 1.6) showed good correlation· compared to the soil-estimated parameters (aJ = 11.04 and ao = 1.02). The factorysupplied parameters gave a linear regression coefficient (r2 ) of 0.822 and root mean square error (RMSE) of 0.062. The soil-estimated parameter showed a linear regression coefficient of 0.820 with RMSE of 0.085. The comparison between the soil water content measured using the PR1 profile probe and Diviner 2000 showed a linear regression coefficient of 0.947 to 0.964 with a range of RMSE of 0.070 to 0.109 respectively for the first 100 to 300 mm soil depths. The deeper depths (400, 600 and 1000 mm) showed a linear regression coefficient ofO.716to 0.810 with a range of 0.058 to 0.150 RMSE. These differences between the shallow and deeper depths could be due to soil variability or lack of good contact between the access tube and the surrounding soil. To undertake irrigation scheduling using the PRl profile probe and Diviner 2000, the soil water content limits were determined using field, laboratory and regression equations. The field method was done by measuring simultaneously the soil water content using the PR1 profile probe and soil water potential using a Watermark sensor and tensiometers at three depths (100, 300 and 600 mm) from a 1 m2 bare plot, while the soil dries after being completely saturated. The retentivity function was developed from these measurements and the drained upper limit was estimated to be 0.355 m3 m-3 when the drainage from the pre-wetted surface was negligible. The lower limit was calculated at -1500 kPa and it was estimated to be 0.316 m3m,3. The available soil water content, which is the difference between the upper and lower limit, was equal to 0.039 m3 m,3. In the laboratory the soil water content and matric potential were measured from the undisturbed soil samples taken from the edge of the 1 m2 bare plot before the sensors were installed. Undisturbed soil samples were taken using a core sampler from 100 to 1000 mm soil depth in three replications in 100 mm increments. These undisturbed soil samples were saturated and subjected to different matric potentials between -1 to -1500 kPa. In the laboratory, the pressure was increased after the cores attained equilibrium and weighed before being subjecting to the next matric potential. The retentivity function was then developed from these measurements. The laboratory method moved the drained upper limit to be 0.390 m3 m,3 at -33 kPa and the lower limit be 0.312 m3m-3 at -1500 kPa. The regression equation, which uses the bulk density, clay and silt percentage to calculate the soil water content at a given soil water potential, estimated the drained upper limit to be 0.295 m3m-3at -33 kPa and the lower limit 0.210 m3 m,3 at -1500 kPa. Comparison was made between the three methods using the soil water content measured at the same soil water potential. The fieldmeasured soil water content was not statistically the same with the laboratory and estimated soil water content. This was shown from the paired-t test, where the probability level (P) for the laboratory and estimated methods were 0.011 and 0.0005 respectively at 95 % level of significance. However, it showed a linear regression coefficient of 0.975 with RMSE of 0.064 when the field method was compared with the laboratory method. The field method showed a linear regression coefficient of 0.995 with RMSE of 0.035 when compared with the estimated method. The timing and amount of irrigation was determined using the PR1 profile probe and Diviner 2000. The laboratory measured retentivity function was used to define the fill (0.39 m3 m-3 ) and high refill point (0.34 m3 m-3 ). The soil water content was measured using both sensors two to three times per week starting from May 29 (149 day of year, 2002) 50 days after planting until September 20 (263 day of year) 11 days before harvesting. There were five irrigations and twenty rainfall events. The next date of irrigation was predicted graphically using, the PRl profile probe measurements, to be on 3 September (246 day of year) after the last rainfall event on 29 August (241 day of year) with 8 mm. When the Diviner 2000 was used, it predicted two days after the PRl profile probe predicted date. This difference appeared since the Diviner 2000-measured soil water content at the rooting depth was slightly higher than the PRl profile probe measurements. The amount of irrigation was estimated using two comparable methods (graphic and mathematical method). The amount of irrigation that should have been applied on 20, September (263 day of year) to bring the soil water content to field capacity was estimated to be 4.5 hand 23 mm graphically and 5.23 hand 20 mm mathematically. The difference between these two methods was caused due to the error encountered while plotting the correct line to represent the average variation in soil water content and cumulative irrigation as a function of time. More research is needed to find the cause for the very low soil water content measurements of the PRI profile probe at some depths. The research should be focused on the factors, which could affect the measurement of the PRl profile probe and Diviner 2000 like salinity, temperature, bulk density and electrical conductivity. Further research is also needed to extend the non-linear relationship between the electrical resistance of the sensor and soil water potential up to -200 kPa. This non-linear equation of the Watermark is only applicable within the range of soil water potential between -10 and -100 kPa.
The use of infrared thermometry for irrigation scheduling of cereal rye (Secale cereale L.) and annual ryegrass (Lolium multiflorum Lam.)
(2003) Mengistu, Michael Ghebrekidan.; Savage, Michael John.; Everson, Colin Stuart.
Limited water supplies are available to satisfy the increasing demands of crop production. It is therefore very important to conserve the water, which comes as rainfall, and water, which is used in irrigation. A proper irrigation water management system requires accurate, simple, automated, non-destructive method to schedule irrigations. Utilization of infrared thermometry to assess plant water stress provides a rapid, nondestructive, reliable estimate of plant water status which would be amenable to larger scale applications and would over-reach some of the sampling problems associated with point measurements. Several indices have been developed to time irrigation. The most useful is the crop water stress index (CWSI), which normalizes canopy to aIr temperature differential measurements, to atmospheric water vapour pressure deficit. A field experiment was conducted at Cedara, KwaZulu-Natal, South Africa, to determine the non-water-stressed baselines, and CWSI of cereal rye (Secale cereale L.) from 22 July to 26 September 2002, and aImual (Italian) ryegrass (Lolium multiflorum Lam.) from October 8 to December 4, 2002, when the crops completely covered the soil. An accurate measurement of canopy to air temperature differential is crucial for the determination of CWSI using the empirical (Idso et al., 1981) and theoretical (Jackson et al., 1981) methods. Calibrations of infrared thermometers, a Vaisala CS500 air temperature and relative humidity sensor and thermocouples were performed, and the reliability of the measured weather data were analysed. The Everest and Apogee infrared thermometers require correction for temperatures less than 15 QC and greater than 35 QC. Although the calibration relationships were highly linearly significant the slopes and intercepts should be corrected for greater accuracy. Since the slopes of the thermocouples and Vaisala CS500 air temperature sensor were statistically different from 1, multipliers were used to correct the readings. The relative humidity sensor needs to be calibrated for RH values less than 25 % and greater than 75 %. The integrity of weather data showed that solar irradiance, net irradiance, wind speed and vapour pressure deficit were measured accurately. Calculated soil heat flux was underestimated and the calculated surface temperature was underestimated for most of the experimental period compared to measured canopy temperature. The CWSI was determined using the empirical and theoretical methods. An investigation was made to determine if the CWSI could be used to schedule irrigation in cereal rye and annual rye grass to prevent water stress. Both the empirical and theoretical methods require an estimate or measurement of the canopy to air temperature differential, the non-waterstressed baseline, and the non-transpiring canopy to air temperature differential. The upper (stressed) and lower (non- stressed) baselines were calculated to quantify and monitor crop water stress for cereal rye and annual ryegrass. The non-water-stressed baselines were described by the linear equations Te - Ta = 2.0404 - 2.0424 * VPD for cereal rye and Te - Ta = 2.7377 - 1.2524 * VP D for annual ryegrass. The theoretical CWSI was greater than the empirical CWSI for most of the experimental days for both cereal rye and annual ryegrass. Variability of empirical (CWSI)E and theoretical (CWSI)T values followed soil water content as would be expected. The CWSI values responded predictably to rainfall and irrigation. CWSI values of 0.24 for cereal rye and 0.29 for annual ryegrass were found from this study, which can be used for timing irrigations to alleviate water stress and avoid excess irrigation water. The non-water-stressed baseline can also be used alone if the aim of the irrigator is to obtain maximum yields. However the non-water-stressed baseline determined using the empirical method cannot be applied to another location and is only valid for clear sky conditions. And the non-water-stressed baseline determined using theoretical method requires computation of aerodynamic resistance and canopy resistances, as the knowledge of canopy resistance, however the values it can assume throughout the day is still scarce. The baseline was then determined using a new method by Alves and Pereira (2000), which overcomes these problems. This method evaluated the infrared surface temperature as a wet bulb temperature for cereal rye and annual ryegrass. From this study, it is concluded that the infrared surface temperature of fully irrigated cereal rye and annual ryegrass can be regarded as a surface wet bulb temperature. The value of infrared surface temperature can be computed from measured or estimated values of net irradiance, aerodynamic resistance and air temperature. The non-water-stressed baseline is a useful concept that can effectively guide the irrigator to obtain maximum yields and to schedule irrigation. Surface temperature can be used to monitor the crop water status at any time of the day even on cloudy days, which may greatly ease the task of the irrigator.
Modelling the soil water balance and applications using a decision support system (DSSAT v3.5).
(2003) Ghebreab, Tesfalidet Alem.; Savage, Michael John.
Water is a scarce resource used by various stakeholders. Agriculture is one of the users of this resource especially for growing plants. Plants need to take up carbon dioxide to prepare their own food. For this purpose plants have stomatal openings. These same openings are used for transpiration. Quantifying transpiration is important for efficient water resource management and crop production because it is closely related to dry matter production. Transpiration could be measured using a number of methods or calculated indirectly through quantification of the soil water balance components using environmental instruments. The use of models such as the Decision Support System for Agrotechnology Transfer (DSSAT v3.5) is, however, much easier than environmental instruments. Nowadays, with increased capabilities of computers, the use of crop simulation modelling has become a common practice for various applications. But it is important that models, such as DSSAT v3.5, be calibrated and verified before being used for various applications such as long-term risk assessment, evaluation of cultural practices and other applications. In this study the model inputs have been collected first Then the model was calibrated and verified. Next sensitivitY analysis was carried to observe the model behavior to changes in inputs. Finally the model has been applied for long-term risk assessment and evaluation of cultural practices. In this study, the data collected formed the basis forthe minimum dataset needed for running the DSSAT v3.5 model. In addition, the factory given transmission of shading material over a tomato crop was compared to actual measurements. Missing weather data (solar irradiance, minimum and maximum air temperature and rainfall) were completed after checking that it was homogeneous to measurements from nearby automatic weather station. It was found that factory-given transmission value of 0.7 of the shade cloth was different from the actual one of 0.765. So this value was used for conversion of solar irradiance measured outside the shade cloth to solar irradiance inside the shade cloth. Conventional laboratory procedures were used for the analysis of soil physical and chemical properties. Soil water content limits were determined using texture and bulk density regression based equations. Other model inputs were calculated using the DSSAT model. Crop management inputs were also documented for creation of the experimental details file. The DSSATv3.5 soil water balance model was calibrated for soil, plant and weather conditions at Ukulinga by modifying some of its inputs and then simulations of the soil water balance components were evaluated against actual measurements. For this purpose half of the data available was used for calibration and the other half for verification. Model simulations of soil water content (150 to 300 mm and 450 to 600 mm) improved significantly after calibration. In addition, simulations of leaf area index (LA!) were satisfactory. Simulated evapotranspiration (E1) had certain deviations from the measured ET because the latter calculated ET by multiplying the potential ET with constant crop multiplier so-called the crop coefficient. Sensitivity analysis and long-term risk assessments for yield, runoff and drainage and other model outputs were carried out for soil, plant and weather conditions at Ukulinga. For this purpose, some of the input parameters were varied individually to determine the effect on seven model output parameters. In addition, long-term weather data was used to simulate yield, biomass at harvest, runoff and drainage for various initial soil water content values. The sensitivity analysis gave results that conform to the current understanding of the soil-plant atmosphere system. The long-term assessment showed that it is risky to grow tomatoes during the winter season at Ukulinga irrespective of the initial soil water content unless certain measures are taken such as the use of mulching to protect the plants from frost. The CROPGRO-Soya bean model was used to evaluate the soil water balance and gro'W1:h routines for soil, plant and weather conditions at Cedara. In addition, cultural practices such as row spacing, seeding rate and cultivars were also evaluated using longterm weather data. Simulations of soil water content were unsatisfactory even after calibration of some of the model parameters. Other model parameters such as LAI, yield and flowering date had satisfactory agreement with observed values. Results from this study suggest that the model is sensitive to weather and cultural practices such as seeding rates, row spacing and cultivar maturity groups. The general use of decision support systems is limited by various factors. Some of the factors are: unclear definition of clients/end users; no end user input prior to or during the development of the DSS; DSS does not solve the problems that the client is experiencing; DSS do not match their decision-making style; producers see no reason to change the current management practices; DSS does not provide benefit over current decision-making system; limited computer ownership amongst producers; lack of field testing; producers do not trust the output due to the lack of understanding of the underlying theories of the models utilized; cannot access the necessary data inputs; lack of technical support; lack of training in the development ofDSS software; marketing and support constraints; institutional resistances; short shelf-life of DSS software; technical constraints, user constraints and other constraints. For successful use of DSS, the abovementioned constraints have to be solved before their useful impacts on farming systems could be realized. This study has shown that the DSSAT v3.5 model simulations of the soil water balance components such as evapotranspiration and soil water content were unsatisfactory while simulations of plant parameters such as leaf area index, yield and phonological stages were simulate to a satisfactory standard. Sensitivity analysis gave results that conform to the current understanding of the soil-plant -atmosphere system. Model outputs such as yield and phonological stages were found to sensitive to weather and cultural practices such as seeding rates, row spacing and cultivar maturity groups. It ha been further investigated that the model could be used for risk assessment in various crop management practices and evaluation of cultural practices. However, before farmers can use DSSAT v3.5, several constraints have to be solved.
The long-term measurement of total evaporation over Acacia mearnsii using large aperture scintillometry
(2007) Clulow, Alistair David.; Savage, Michael John.
A large aperture scintillometer (LAS) was operated continuously over a distance of 575 m from 19 August 2006 to 29 September 2007 in the South African KwaZulu-Natal midlands mistbelt area over Wattle (Acacia mearnsii). The LAS measurements of the structure parameter of the refractive index of air ( 2 n C ), were used to calculate the sensible heat flux. The shortened energy balance equation was used to estimate the latent energy flux as a residual from which the total evaporation (ET) was calculated. The LAS estimates of sensible heat flux during the short transition period (1 hour) between stable and unstable conditions were on occasion erroneous and required verification. Advection was also found to affect sensible heat flux estimates. Long-term operation of the LAS was however found to be possible even at remote sites, producing reliable and continuous results. The LAS estimates of sensible heat are sensitive to zero-plane displacement height and wind speed data impact and these should be derived as accurately as possible. Tree heights were measured at monthly intervals and a zero-plane displacement and effective height were calculated every two weeks. The sensible heat flux was thus processed in two week blocks of data corresponding to progressive effective heights. The tree growth rate was consistent over time and was not affected by seasonality, indicating that reduced air temperatures, rainfall and solar irradiance in winter are not limiting growth. The average growth rate was 0.37 m per month or 4.5 m per year. The LAS ET was compared to the American Society of Civil Engineers - Environmental and Water Resources Institute (ASCE-EWRI) short grass reference evaporation (ETsz) for a seven-month period and was found to compare favourably (R2 = 0.78) with outliers caused by advection and rainfall events. Calculations of grass reference evaporation at hourly and daily intervals provide different results. The daily estimates are lower than the hourly estimates by 17 % on average. Where hourly data is summed to calculate a daily ETsz, night-time values should be included. The LAS ET measurements were validated against the Priestley and Taylor (1972) method of estimating ET and found to be in good agreement (R2=0.94). The Priestley and Taylor daily total latent energy flux, from 22 August 2006 to 29 September 2007, was 9 % higher than the LAS results on average. The Bowen ratio for the entire period is less than 1, indicating that the latent energy flux dominates at the site. The ET over the period of measurement (13 months) is 1250 mm and the rainfall is 750 mm. This confirms previous results at the site using the Bowen ratio energy balance method showing that the ET exceeds the rainfall by 45 % and justifies further research into soil water, ground water and root interactions in the deep soil profile.
Seasonal variation of surface energy fluxes above a mixed species and spatially homogeneous grassland.
(2011) Moyo, Nicholas C.; Savage, Michael John.
The increasing human population, industrialization, urbanisation and climate change challenges have resulted in an increased demand for already scarce water resources. This has left the agricultural sector with less water for production. Sustainable water management strategies would therefore require accurate determination of water-use. In agriculture, water-use can best be determined from total evaporation which is the loss of water from soil and vegetation to the atmosphere. Accurate quantification of total evaporation from vegetation would require a thorough understanding of water transport processes between vegetation and the atmosphere, especially in a water-scarce country like South Africa. Several methods for estimating total evaporation have been developed and are in use today. Some of the common methods used today are: the Bowen ratio energy balance, eddy covariance, scintillometry, flux variance and surface renewal. However, various methods have advantages and disadvantages. Considerations include the cost of equipment and level of skill required for use of some of the methods. A number of methods involve indirect or direct estimation of sensible heat flux then calculating latent energy flux and hence total evaporation as a residual of the shortened energy balance equation. The main objective of this study is to determine the effects of grassland management practices on the energy balance components as well as on the surface radiation balance. Eddy covariance and surface renewal methods were employed to investigate the effects of grassland management practices (mowing and burning) on the micrometeorology of naturally occurring grassland. A 4.5-ha grassland site (Ukulinga, Pietermaritzburg, South Africa) was divided into two halves: one area was initially mowed (cut-grass site) to a height of 0.1 m while the other was not mowed (tall-grass site). The tall-grass site was later treated by burning and hence referred to as the burnt-grass site. Two eddy covariance systems were deployed, one at each of the cut-grass and the tall-grass sites. The systems each comprised a three-dimensional sonic anemometer to measure high frequency sonic temperature, orthogonal wind speeds and directions and the eddy covariance sensible heat flux (W m-2). Latent energy flux, from which total evaporation was then determined, was calculated as a residual from the shortened energy balance equation from measurements of sensible heat flux, net irradiance and soil heat flux assuming closure is met. Other microclimatic measurements of soil water content, soil temperature, surface reflection coefficient and reflected solar irradiance were performed, the latter with a four-component net radiometer. An automatic weather station was also set up at the research site for continuous measurements of solar irradiance, air temperature, relative humidity, wind speed and direction and rainfall. Water vapour pressure and grass reference evaporation were also determined online. Energy fluxes from the tall-grass site were measured from March to June 2008. Greater total evaporation rates (2.27 mm day-1) were observed at the beginning of the experiment (March). As winter approached most of the energy balance components showed a constant decreasing trend and the average total evaporation rates for May and June were 1.03 and 0.62 mm day-1, respectively. The tall-grass site had consistently lower soil temperatures that changed diurnally when compared to the cut-grass site. The soil water content at both sites showed no significant differences. Most of the energy balance components were similar between the two sites and changed diurnally. Although there were small differences observed between other energy balance components, for example, latent energy flux was slightly greater for the tall-grass site than for the cut-grass site. The tall-grass site had more basal cover and this may have contributed to the differences in temperature regimes observed between the two sites. However, the plants growing at the cut-grass site showed more vigour than the ones at the tall-grass site as spring approached. Burning of a mixed grassland surface caused significant changes to most of the optical properties and energy fluxes of the surface. Following burning, the soil temperature was elevated to noticeable levels due to removal of basal cover by burning. The surface reflection coefficient measured before and after the burn also presented a remarkable change. The surface reflection coefficient was significantly reduced after the burn but a progressive increase was observed as the burnt grass recovered after the spell of spring rains. The energy fluxes: net irradiance, latent energy flux and soil heat flux also increased following the burn but the latent energy flux was reduced as transpiration was effectively eliminated by the burning of all actively transpiring leaves. As a result, the main process that contributed towards latent energy flux was soil evaporation. An ideal surface renewal analysis model based on two air temperature structure functions was used to estimate sensible heat flux over natural grassland treated by mowing. Two air temperature lag times r (0.4 and 0.8 s) were used when computing the air temperature structure functions online. The surface renewal sensible heat fluxes were computed using an iteration process in Excel. The fluxes, obtained using an iterative procedure, were calibrated to determine the surface renewal weighting factor (a) and then validated against the eddy covariance method using different data sets for unstable conditions during 2008. The latent energy flux was computed as a residual from the shortened energy balance equation. The surface renewal weighting factor was determined for each of the two heights and two lag times for each measurement height (z) above the soil surface. The a values obtained during the surface renewal calibration period (day of year 223 to 242, 2008) ranged from 1.90 to 2.26 for measurement height 0.7 m and r = 0.4 and 0.8 s. For a measurement height of 1.2 m and r = 0.4 and 0.8 s, a values of 0.71 and 1.01 were obtained, respectively. Good agreement between surface renewal sensible heat flux and eddy covariance sensible heat flux was obtained at a height of 1.2 m using a = 0.71 and a lag time of 0.4 s. Total evaporation for the surface renewal method was compared against the eddy covariance method. The surface renewal method, for a height of 1.2 m and a lag time of 0.4 s, yielded 1.67 mm while the eddy covariance method yielded 1.57 mm for a typical cloudless day. For the same day for a measurement height of 1.2 m and a lag time of 0.8 s, eddy covariance and surface renewal methods yielded 1.57 and 1.10 mm, respectively. For a lag time of 0.4 s, the surface renewal method overestimated total evaporation by 0.10 mm while for a lag time of 0.8 s, the total evaporation was underestimated by 0.47 mm. As a result, the surface renewal method performed better for z = 1.2 m and a lag time of 0.4 s. The eddy covariance method gave reliable sensible heat fluxes throughout the experiment and this allowed a comparison of fluxes across all treatment areas to be achieved. The short-term analysis of the surface renewal method also gave reliable energy fluxes after calibration. Compared to the eddy covariance method, the surface renewal method is more attractive in the sense that it is easy to operate and use and it is relatively cheap. However, the surface renewal method requires calibration and validation against a standard method such as the eddy covariance method. This study showed that grassland management practices had a considerable effect on surface radiation and energy balance of the mowed and burnt treatment sites. Total evaporation was mainly controlled by the available energy flux, rainfall and grassland surface structure. High total evaporation values were observed during summer when net irradiance was at its highest and grass growth at its peak. Low total evaporation values were observed in winter (dry atmospheric conditions) when net irradiance was at its lowest and most vegetation was dormant.
Monitoring fire danger in near real-time using field-based agrometeorological measurement systems.
(2014) Strydom, Sheldon.; Savage, Michael John.
Africa has been termed the ‘Fire Continent’ due to its high annual fire frequency. Wildfires are considered one of the most common disasters in South Africa resulting in a high number of human fatalities and financial loss on an annual basis. It is believed that increased population growth, as well as more concentrated settlement planning is likely to result in increased fire disasters and increased human fatalities as a direct result of wildfires. The high number of human fatalities and high financial loss associated with wildfires served as the main motivation for the research throughout all studies. While wildfires may provide beneficial environmental service, increased wildfire activity can result in a number of adverse effects on the environment, for example the removal of vegetation, fascilitating/aggravating floods and soil erosion but do bring with them positive effects such as nutrient recycling and removal of alien species. In order to better understand the spatial and temporal variations and characteristics of wildfires in South Africa an 11-year dataset of MODIS-derived Active Fire Hotspots was analysed using an open source geographic information system. The study included the mapping of national fire frequency over the 11-year period. Results indicate that the north-eastern regions of South Africa experience the greatest fire frequency, in particular the mountainous regions of KwaZulu-Natal, Mpumalanga and the Western Cape. Increasing trends in provincial fire frequency was observed in eight out of the nine provinces with Mpumalanga being the only province where a decrease in annual fire frequency over the study period was observed. Temporally, fires have been observed in all months for all provinces although distinct fire seasons were observed, largely driven by rainfall seasons. The South-Western regions of South Africa (winter rainfall patterns) experienced higher fire frequencies during the summer months with the rest of the country (summer rainfall) experiencing higher fire frequencies during the winter months. Regions which experience bi-modal rainfall seasons did not display distinct fire seasons. The study included an investigation into the likely effects of climate change on South African fire frequency. Three of the 11 years were identified as being climatologically anomalous. Fire frequencies in 2005 and 2010 (two of the warmest years in recent history) were significantly greater than normal years. Observed fire frequencies in 2008 were also significantly greater. The increased fire frequency was attributed to a severe La Niña event which may have resulted in increased vegetation growth prior to the dry season. A current issue with the mitigation of wildfires is the lack of proper real-time monitoring and measurement systems which can aid decision makers in the timing of controlled fires. The development of a system for improved monitoring of meteorological conditions conducive to fire was investigated. The traditionally used nomogram and lookup table used by Lowveld fire danger index (LFDI) system was replaced by mathematical functions which were then programmed into an automatic weather station datalogger. Near real-time results of the calculated LFDI were displayed in a web-based teaching, learning and research system found at: http://agromet.ukzn.ac.za:5355/?command=RTMC&screen=Fire%20danger%20index. Warm, dry mountain winds known as Berg winds have a direct link to fire weather, enhancing the danger of uncontrollable fires. Berg winds are associated with periods of increased air temperature, decreased relative humidity and increased wind speeds. This increases the potential for the development and spread of wildfires. The effects of Berg winds on the microclimate and fire danger were quantified. For this purpose, historic hourly meteorological data, local and international, were used together with a fuzzy logic system for determining Berg wind conditions in near real-time. This included the use of modelled diurnal sinusoidal functions for solar irradiance, air temperature, relative humidity, wind speed and direction, for various locations. Application of the system demonstrated that out of four sites in the KwaZulu-Natal Midlands, South Africa, Baynesfield was the most vulnerable to Berg wind occurrences, followed by Ukulinga (near Pietermaritzburg) and lastly Cedara. Boulder in Wyoming, USA, experienced a larger frequency of Berg wind events compared to Sidney in Montana, USA, despite being located at a higher altitude. The McArthur Fire Danger Meters (FDM) have been used to monitor and measure fire danger in Australia since the 1950s when they were first developed following severe wildfires. The McArthur FDM has traditionally made use of nomograms to calculate grassland and forest fire danger. In the 1980s mathematical descriptions of the McArthur FDM were developed. These then allowed for the sub-daily and near real-time measurement of grassland and forest fire danger. While commercial forestry contributes only a small percentage of South Africa’s GDP, forest fires have been identified as a common occurrence in commercial forestry plantations and natural forests, which can at times spread to neighbouring grasslands or non-forest vegetation. Forest fire disasters in South Africa are well documented by the mainstream media and it is believed that proper monitoring of forest fire danger, along with grassland fire danger, may mitigate potential wildfire disasters and limit human fatalities and financial loss. The McArthur FDM also provides opportunity to model fire behaviour – a component of wildfire mitigation lacking in the Lowveld fire danger index. The study utilized the equations developed to investigate the use of the McArthur FDM in South Africa by applying the equations to historical hourly meteorological data recorded at four locations in the KwaZulu-Natal Midlands, South Africa. Modifications to the McArthur grassland fire danger index (GFDI) were needed and resulted in the development of a method to calculate vegetation curing as a function of micrometeorological variables. Datasets were analysed to investigate seasonal curing at the four locations with results indicating high monthly curing averages in all months, with the greatest range of curing averages experienced in the winter months. Frequency analysis of both the GFDI and the forest fire danger index (FFDI) indicate that lower fire danger ratings are more common than high danger ratings as one would expect. The GFDI and FFDI displayed acceptable sensitivity to changes in the microclimate as observed through conducting sensitivity analysis and by plotting diurnal variations of GFDI and FFDI during Berg wind events. The fire behaviour modelling component of the McArthur FDM does not yield realistic results on steep topography but does provide a baseline on fire characteristics on gentler slopes
Modelling the impacts of increased air temperature on maize yields in selected areas of the South African highveld using the cropsyst model.
(2014) Pasi, Jonathan M.; Savage, Michael John.
Abstract available in PDF file.
Web-based teaching, learning and research using real-time data from field-based agrometeorological measurement systems.
(2014) Savage, Michael John.; Everson, Colin Stuart.
Abstract available in PDF file.
Irrigation control system with a web-based interface for the management of Eucalyptus planting stock.
(2015) Kaptein, Nkosinathi David.; Savage, Michael John.
Commercial forestry nurseries use large quantities of water to irrigate the planting stock to meet the annual forestry industry planting demands. However, South Africa is a water scarce country and there is high competition for this limited resource with other sectors. Thus, sustainable water management strategies should be put in place to preserve this precious resource. In commercial forestry nurseries, sustainable water use may be achieved by carefully managing irrigation schedules, such as accurately measuring growing media water content and then replenishing the depleted soil water. Improved nursery irrigation management may not only save water, but also has the potential to reduce the prevalence of pests and diseases, excessive leaching of nutrients, irrigation costs and may produce robust planting stock that is better suited to adapt to field conditions. Growing media water content can be directly measured by the gravimetric method. However, this method is laborious, time consuming, costly and does not allow for near real-time monitoring and control. Several indirect methods for estimating soil water content have been developed and are in use today. Some common methods are: frequent domain reflectometry, time domain reflectometry, time domain transmission and the dual-needle heat pulse method. However, each method has advantages and disadvantages. Considerations for choosing the most appropriate method are the ability to automate, accuracy and precision, skills required to use and the costs to purchase. The main objective of this study was to calibrate the low cost commercially available Decagon EC-5 soil water content sensors using nursery growing media. The calibration equation was then used to measure and control irrigation for Eucalyptus grandis x Eucalyptus urophylla and Eucalyptus dunnii planting stock grown in seedling trays in a greenhouse. A web-based data and information system was utilised to share measurements and display greenhouse environmental conditions in near real-time. The data could be viewed or downloaded using the internet1. Decagon EC-5 soil water content sensors were laboratory-calibrated, using nursery growing media, against the standard gravimetric method. Four nursery growing media were used for calibrations: coir/perlite mix (CP), coir/pine bark/vermiculite mix (CPBV), pine bark (PB) and sandy soil. The calibration relationships between gravimetric water content and sensor output for each growing media were established, and the manufacturer supplied calibration equation was evaluated against laboratory calibration equations. The appropriate laboratory calibration equation was programmed in the datalogger to measure and control irrigation. Greenhouse microclimate measurements of air temperature, relative humidity and solar irradiance were conducted. Hourly grass reference evaporation (ETo) was calculated by the datalogger. The greenhouse microclimate measurements were compared against an open area automatic weather station at the University of KwaZulu-Natal Agrometeorology Instrumentation Mast (UKZN AIM) system measurements. The EC-5 soil water content sensors were used to schedule irrigation for E. grandis x E. urophylla hybrid clones (GxU) (Experiment 1) and E. dunnii seedlings (Experiment 2) grown in seedling trays inside a fully air temperature controlled greenhouse. Irrigation was controlled in three treatments: low, medium and high watering. All treatments were treated the same except for differing irrigation application. Total daily irrigation and drainage were measured per treatment. Irrigation scheduling for GxU was programmed at a set point and E. dunnii seedlings at lower and drained upper limits. Seedling measurements conducted were root collar diameter (RCD), heights, stomatal conductance (gs), root-to-shoot ratio and total leaf area. Total drainage and its electrical conductivity (EC) was also measured. The calibration relationship showed a linear relationship between gravimetric water content and sensor output for all four growing media with an R2 greater than 0.92. The manufacturer supplied calibration equation poorly estimated growing media water content compared to the laboratory calibration. Poor estimation exceeded the 5% error specified by the manufacturer. Air temperature was consistently less than 25°C inside the greenhouse. The external air temperature, as measured by the UKZN AIM system, fluctuated and reached a maximum of 36.6°C during the study period. Solar irradiance inside the greenhouse was 60% lower than that measured by the UKZN AIM system. The relative humidity was consistently higher inside the greenhouse compared to that measured by the UKZN AIM system. Greenhouse grass reference evaporation was consistently lower than the UKZN AIM system due to low air temperature and high RH inside the greenhouse. For Experiment 1, the GxU clones were irrigated too frequently for short periods. This led to over- and under- irrigation in high and low watering treatments, respectively. However, these challenges were addressed in Experiment 2 using E. dunnii seedlings irrigated at lower and drained upper irrigation limits. In Experiment 2, variability in sensor measurements within each treatment were observed at drained upper limit and decreased at lower limit. This was likely caused by a change in the pore space volume from dry to wet growing media. The web-based system was successfully used as an early warning system to monitor soil water content measurements and greenhouse microclimate, averting experimental failure due to lack of irrigation on one occasion. Seedlings in the high watering treatment had the highest RCD, heights and gs followed by the medium and low watering treatments. Although the low watering treatment had the lowest growth rates and gs these seedlings were more robust, hardy and resistant to water stress. The root-to-shoot ratio showed no statistically significant differences between treatments. However, seedlings in the high watering treatment had slightly greater root volume. This was probably due to the increased total seedling leaf area for this treatment which facilitated increased photosynthetic activity and carbohydrates production, enabling increased root growth. The highest EC measurements were recorded in the low watering treatment. This was likely due to low irrigation and therefore nutrients were not washed off the growing media. Medium watering treatment EC was 30% lower than the low watering treatment whilst high watering EC was almost equivalent to the irrigation water. The analysis of economics showed that implementing the fully automated system could be costly. However, there are many potential benefits that may be offered by this system such as reduction in water use, pumping costs and management time. The early warning offered by this system could potentially help avoid the loss of planting stock if there is a problem with the irrigation system. The study showed that irrigation may be automatically scheduled for nursery seedlings trays using low cost Decagon EC-5 soil water content sensors with reasonable accuracy. However, medium-specific calibration is important to improve the soil water content measurement accuracy. The study also showed that reducing irrigation may result in reduced growth rates of seedlings. However, other benefits such as seedling resistance to water stress, robust seedlings, irrigation water savings and a reduction in washing off nutrients may be achieved.
Investigation of microclimate for human comfort in the natural environment and in a car parked in the open.
(2016) Luthuli, Sithandiwe Ignatia.; Savage, Michael John.
Abstract available in PDF file.
Microclimate modification to improve productivity of ‘Carmen®-Hass’ avocado orchards using shadenet under subtropical conditions of Limpopo Province, South Africa.
(2016) Malapana, Clarance Kgethego.; Savage, Michael John.
The agricultural environment is a complex and dynamic system. Microclimate, the crop, biosphere, and management practices interact to determine the best yield production. South Africa is a water-scarce country, with high variability in annual rainfall. Thus, water quality and quantity are major limiting factors in agriculture. Hence, shadenetting can be used to modify the orchard microclimate to make the environment more conducive for fruit production. The South African avocado industry is export-oriented, so there is a commercial need to optimise the exportable percentage of avocado fruit. Sunburn, wind and hail damage and small fruit size as a result of water stress are the major cull factors for the industry. It is believed that shadenetting, with changes in management practices, can counter these limiting factors. There is no literature on growing avocado fruit under shadenetting. Therefore, the aim of the research was to determine the effects of a 20% white shadenet on ‘Carmen®-Hass’ avocado orchards and productivity. The long term objective is to improve avocado fruit quality and profitability in the Mooketsi Valley, Limpopo province, South Africa, a subtropical environment by reducing abiotic stress, particularly, solar irradiance, heat and wind. The trial was conducted at Goedgelegen Estate in the Mooketsi Valley on ‘Carmen®-Hass’ trees planted in 2007/8 season. A 1-ha shadenet structure (6 m high) was used, with 20% white shadenet over the roof and 40% green shadenet on the sides. Air and canopy temperature, relative humidity, wind speed, solar irradiance and leaf wetness duration (LWD) and sap flow were monitored at a sub-hourly rate. Evapotranspiration was calculated from the above mentioned parameters. Irrigation was monitored five times per week using tensiometers at 300- and 600-mm soil depths. The comparison between open and shadenet leaf areas showed that leaves in the open treatment were reduced as a result of the abiotic stress. Fruit water content under the shadenet compared to the open was greater, such enabled fruit under the shadenet to reach maturity two weeks earlier when compared with open treatment. Air and canopy temperature and relative humidity were slightly reduced under the shadenet, with the greatest difference occurring during the flowering period in mid-winter. The modification in air temperature and relative humidity was beneficial for bee activity and pollination in 2015 compared to the 2014 season. ‘Carmen®-Hass’ flowers in mid-winter when temperature conditions are not conducive for pollination. Canopy temperature was also reduced under shadenetting compared to the open treatment. The reduction was due to differences in tree density and the role that shadenetting plays. The infrared thermometer measurements were uniform with dense canopies compared to sparse tree canopies. The midday incoming solar irradiance was reduced by 18% under the shadenet compared to the open treatment. Calm conditions were experienced under the shadenet. Hence, windspeed was reduced to negligible levels. Also, the shadenet resist air flow to a certain height compared to the open treatment. LWD was extended by 12% under shadenet. An infestation of the insect pest citrus leaf roller (Archips occidentalis) caused severe damage to the fruit during the 2014/15 season due to the high plant density used. Significant results were that evapotranspiration was reduced by 14 and 29% less water was applied under the shadenet to maintain an adequate soil water content compared to the open treatment. Fruit reached minimum maturity two weeks earlier under shadenet compared to the open treatment. Fruit quality and pack-out were improved under the shadenet due to reduction in sunburn, wind damage and small fruit. But poor yields were experienced during the 2014 season due to poor bee activity, pollination and fruit size distribution were reduced under the shadenet compared to the open treatment. But following the improved bee activity in 2015, the 2016 normal season yield is likely to be improved under the shadenet than in the open treatment. Data collected in the Mooketsi Valley showed that 20% white shadenet has modified the microclimate and improved fruit quality. The water use under the shadenet was improved compared to the open treatment. But a thorough investigation on bee management under shadenet is required to optimise pollination in order to obtain greater yields under the shadenet.
The radiation balance of Midmar Dam in KwaZulu-Natal, South Africa.
(2016) Myeni, Lindumusa.; Savage, Michael John.
Abstract available in PDF file.
Investigating possible impact of climate change on sugarcane production in KwaZulu-Natal, South Africa.
(2016) Sithole, Bonga Benson.; Savage, Michael John.
The KwaZulu-Natal Department of Agriculture, Environmental Affairs and Rural Development in 2010 undertook to investigate the vulnerability of KwaZulu-Natal to climate change which identified various sectors of the economy in the region that are impacted by climate change. The aim of the current study was to investigate the possible impact of climate change on sugarcane production in KwaZulu-Natal. The main objective of the study was to identify and synthesize current knowledge, scientific literature and data relating to specific aspects of climate change in KwaZulu-Natal, South Africa. In order to achieve the objectives of the study, a questionnaire was developed to ascertain sugarcane farmers’ awareness about climate change. Based on questionnaire data, the study reveals that sugarcane farmers are aware of the effect of climate change on sugarcane production. Cane growers are also aware of their activities that contribute to climate change. Climatological data for the region were collected and analyzed through the Decision Support System Agro-tech Transfer (DSSAT) daily crop model to assess possible climate change impacts on sugarcane production. Daily rainfall, solar radiation, wind-speed, air temperature (minimum and maximum) and dew point temperature data for various sugarcane mill supply areas were collected from the South African Sugar Research Institute. RClimDex (1.0) software was used to determine if climate change did occur for the period 1966 to 2016. The climate data sets were positive to climate change with respect to daily maximum air temperature, daily minimum air temperature, daily rainfall, daily dew point temperature and daily solar radiation. The impact of climate change on sugarcane production in KwaZulu-Natal has been recognized as the main cause for yield reduction. No major decline in sugarcane production has been noted in KwaZulu-Natal for those farmers practicing irrigation and improved management. The study also demonstrates an increase in the amount of trash on sugarcane in the latter years of the study period. In general, the approach presented in this study encompassed and assessed the effect of climate change on sugarcane production with inclusion farmers perception can be considered as a strategic issue on existing climate change concern in KwaZulu-Natal province. Further research about what measures have been implemented by other countries in addressing climate change is recommended.
Rainy season characteristics with reference to maize production for the Luvuvhu river catchment, Limpopo Province, South Africa.
(2017) Tshililo, Fhulufhelo Phillis.; Savage, Michael John.
In arid and semi-arid regions, crop yields are mainly dependent on the amount and spatio-temporal distribution of rainfall. For most smallholder farmers in rural areas of southern Africa, rainfall is a critical input to agricultural production of most staple crops such as maize. To effectively plan for agricultural development, it is of utmost importance that the spatial distribution and temporal variation of rainfall is understood as it govern the type of farming systems that can be practiced in any region. Therefore, a detailed understanding of rainfall is necessary before any farming activities can commence. The study investigated rainy season characteristics with reference to maize production in the Luvuvhu River Catchment. Rainy season characteristics assessed included aridity index, onset, cessation, length of the season, false onset, dry spells, seasonal rainfall, number of rainy days and monthly rainfall. Historical daily rainfall and minimum and maximum air temperature data (1923-2015) were obtained from the Agricultural Research Council. Twelve meteorological stations that were evenly distributed and represented different climatic regions within the catchment were chosen. An aridity index for different areas of the catchment was calculated using the United Nations Environment Programme equation. Evapotranspiration was calculated using the Hargreaves and Samani equation. Annual rainfall was calculated by summing daily rainfall from 1st January to 31st December. The Instat+ v 3.36 statistical programme was utilized to calculate onset, cessation, and length of the season, the number of rainy days, dry spells, seasonal rainfall and monthly rainfall. The Statistica software was used to generate descriptive statistics as well as to calculate probability of exceedance and non-exceedance for the rainy season characteristics. The Anderson-Darling goodness of fit test was performed to determine the distribution model that best represents the data. The resultant probabilities of exceedance were then computed from the distribution models that best fit the data. A non-parametric Spearman rank correlation coefficient test was used to analyze data for trends in rainy season characteristics as well as monthly rainfall. The results from the study showed that monthly rainfall at the Luvuvhu River Catchment during the rainy season varies temporally and spatially. In the high rainfall areas of the catchment, the rainy season commences early from the third week of October and ends the first week of April the following year. For dry areas of the catchment, the rainy season commences in the second week of November and ends early in the third week of February. The results further show a decrease in length of the rainy season, the number of rainy days, and seasonal rainfall further away from wet to dry areas of the catchment. There was no significant change on the onset of the rainy season on the catchment for the past 27-90 years. There is a high risk of both short and medium dry spells at most stations during the month of October, with, Folovhodwe, Phafuri and Sigonde being at highest risk. Farmers are therefore advised to use the first onset for land preparation and plant after the second onset in November and December to avoid the high risk of dry spells and false onset in October and November, depending on the location at the catchment. Folovhodwe, Mampakuil, Phafuri and Sigonde have a mean length of rainy season of less than 120 days and seasonal rainfall of less than 500 mm per rainy season. Hence, these areas are not suitable for rain-fed maize in the current climate. However, they are suitable for the production of other crops which may be sold in order to purchase maize. The most favourable sites for maize production within the catchment are Entabeni, Levubu, Lwamondo, Thathe, Tshiombo, and Vreemedeling. Therefore, production should be maximized at these areas so that there is sufficient maize for the whole catchment. In dry years, stations situated in the low-lying areas in the north-eastern and eastern parts of the catchments receive less rainfall which does not permit planting of maize. In normal and wet years, rainfall is sufficient for the production of various crops. However, in semi-arid areas of the catchment, plans should be made for supplementary water due to high evapotranspiration rates in order to maximize maize production. Stations in the middle/south western parts of the catchment can receive significant rainfall in both dry, normal and wet years. Trend analysis for long-term rainfall data did not show any significant changes in monthly rainfall except for Lwamondo and Levubu where an increasing trend is notable in January rainfall. In December, the rainfall trend was significant at Entabeni, Folovhodwe and Lwamondo. An increase in rainfall is notable at Lwamondo and a decrease in rainfall at Entabeni and Folovhodwe.
Runoff simulation using the SWAT model for flood frequency analysis and design flood estimations in the Luvuvhu River catchment, South Africa.
(2018) Thavhana, Mulalo Precious.; Moeletsi, M'aseapa Mookho Violet.; Savage, Michael John.
Abstract available in PDF file.

Browse

Browsing Masters Degrees (Agrometeorology) by Issue Date