Cover Image
close this bookDDT in the Tropics: The Impact on Wildlife (NRI, 1994)
close this folderPart 2 : Scientific report
close this folderEffects on wildfire
View the document1 Deposition and dissipation
View the document2 Critical soil processes
View the document3 Terrestrial invertebrates
View the document4 Lizards
View the document5 Woodland birds
View the document6 Nocturnal animals

2 Critical soil processes

I F Grant

Litter degradation
Soil respiration

DDT dissipates relatively quickly from surfaces and soils in tropical climates, but it remains a temporary hazard to surface-living and subterranean non-target organisms. By killing soil fauna and altering the community structure it can damage soil fertility in farmland) . Application rates in farmland were generally very high whereas contamination of soil in woodland sprayed with DDT for tsetse fly control is localized. Nonetheless, any effect on key degradative processes will increase the risk of reducing the productivity of soils already low in inherent fertility.

This chapter reports the results of three short-term studies designed to assess the impact of DDT on critical degradative systems that contribute to the maintenance of soil fertility. Rates of decomposition of leaf litter, soil respiration and nitrification were chosen as indicators of the breakdown and mineralization of organic matter.

Litter degradation

The recycling of nutrients is essential for plant production, while soil organic matter influences soil fertility by promoting soil aggregation, increasing water holding and cation exchange capacities, and maintaining microbial biomass. The agents of recycling, micro-organisms and detritivores, are themselves links in food chains.

Dead mopane leaves were put in nylon mesh bags and buried or tethered to the ground in sprayed and unsprayed areas (Box 2.1, Figure 2.1). The bags varied in mesh size to allow different components of the decomposer community access to the leaves. Bags were reweighed after 6 or 17 months.

In the dry season, bags tethered on the surface lost little weight (mean 17% loss, all bags) irrespective of spray treatment or mesh size. However, the rate of litter degradation was significantly reduced at sprayed sites in buried bags accessible to macro-invertebrates. This effect was more marked in the loamy sands of Julbernardia woodland than in silt loam soil, despite more rapid dissipation of DDT residues from the former. There was no evidence DDT affected the breakdown of litter by small invertebrates and microbes, which processed as much as macro-invertebrates in unsprayed areas.

Box 2.1 Litter decomposition

Nylon mesh bags filled with 3 g of dry, uncontaminated mopane leaves were buried to a depth of about 5 cm or tethered to the surface at various sites (n = 12) and left for 6 months (through one wet or one dry season) or for 17 months (two dry seasons and one wet). Several mesh sizes were used to allow access to different components of the decomposer community. In the dry season study, 64,600 and 4000 um mesh bags were used while 125 and 2000 um mesh bags were used at other times.

Surface-tethered litter bags showed weight loss of less than 17% during the dry season irrespective of treatment or mesh size (Figure 2.2). This indicates that dry season microclimatic conditions (low humidity, high temperatures and radiation) at the woodland floor present detritivores and litter harvesters with an inhospitable environment. Nutrient wash-out from early wet season rains was probably the main agency of loss as found in other studies.

Dry season decomposition of litter buried in silty loam soil was most rapid in the largest mesh bags, where up to 60% was lost in 6 months (Figure 2.2). Decomposition rates in the 600 um and 64 um bags were significantly lower, between 30-38% showing that subterranean macro-invertebrate decomposers were active in the dry season or during unseasonable early rains. There were no significant differences between decomposition rates in the smaller mesh bags or between rates at sprayed and unsprayed sites, suggesting that decomposition by microbes, mites and nematodes was unaffected by DDT. However, significantly more litter was removed from 4000 um bags at the unsprayed site (p= <0.5) than at sites sprayed 1-3 times over a three year period suggesting the activity of subterranean macro-invertebrates was reduced by DDT during the dry season.

Litter degradation in buried, large mesh bags was faster in wet (85-90%) than dry season, but rates in the smaller mesh bags were comparable (35-38%) (Figures 2.2,2.3). In silty loam soil losses during the wet season from 2000 um and 125 um mesh bags were similar at sprayed and unsprayed sites (Figure 2.3).

Seventeen months after burial, 20-40% of the litter in 125 um bags had decomposed, irrespective of soil type (Figure 2.4). Rates of decomposition at four unsprayed and five sprayed sites were similar, re-inforcing dry-season findings that DDT does not affect catabolic microbial activity. Over the same period, 89-99% of the litter in 2000 um bags had been decomposed or removed in unsprayed and sprayed areas where soils were classified as silt

loams. However, in loamy sands, supporting Julbernardia woodland, litter disappearance was significantly slower, and at three heavily contaminated sites it was reduced to 16-27%. Although the number of sites was limited, it appeared that not only were detritivorous macro-invertebrates less active or abundant in the sandy soil, but also that the effect of DDT on their activities were more severe in these soils than in silty loam.

Using termite (Trinervitermes trinervoides) bioassays, Wiese demonstrated that soil texture markedly affected the adsorption and subsequent efficacy of DDT, which was only half as effective in a loam as in a sandy soil. The same bioassay showed a gradation in leaching of DDT. It is greatest in sand, less in loam and least in clay. Hence, the retarded rates of litter processing by macro-invertebrates in the loamy sand might be explained by the effect of greater biological activity and downward movement of DDT in these soil types. Quantitative data on macro-invertebrate populations were not collected for sandy soils Processing by microfauna and flora was unaffected by DDT, but overall rates of decay were slower in the loamy sand despite the more rapid dissipation of DDT residues. Fewer bound residues of DDT would be expected in a sandy soil), but as organic matter content in both soil types were comparable, low soil moisture was indicated as the constraint to microbial decomposition.

Soil respiration

Carbon dioxide (CO2) production 1 m from tree trunks was measured in situ at eight sites in the wet season and ex situ in samples of loamy sand and silty loam soils over 350-400 h using a portable infra-red gas analyser. Ex situ samples were mixed with dried grass (0.5% w/w), wetted and incubated to promote microbial and enzyme activity. Soil and air temperatures were recorded during measurement of CO2 efflux to allow temperature-related fluctuations in respiration rates to be identified.

Ex situ respiration rates were generally higher in samples of silty loam soil than in loamy sand and no effects of contamination were apparent (Figure 2.5). In situ rates of CO2 production varied quite widely and were generally higher in less contaminated soils (Table 2.1). However, in sit'' rates can be strongly influenced by root and animal activity. Local plant growth probably explains the highest respiration rate recorded, and low rates would be expected in two of the more contaminated samples on the basis of soil type alone. Although an effect of contamination cannot be ruled out, the natural variability of respiration within sites (CV 20-50%) was not high enough to mask any serious suppression of heterotrophic activity, which has only been seen with very high dosage rates e.g. 50 kg/ha5.

Table 2.1 In situ soil respiration at eight sites in sprayed and unsprayed woodland


Ammonium, produced from the mineralization of soil organic matter, is oxidized by nitrifying bacteria to nitrate. The availability of nitrate is critical for plant growth.

Nitrification rates in sandy clay loam soils were measured using the field method of Grant. Samples were sieved to remove large objects and 100 g aliquots placed in plastic containers and amended with 100 ug ammonium/g dry weight of soil. Deionized water was then added to bring the soils to 70% of field capacity. Just before spraying, the containers were placed in pairs (n = 3) beside large trees in the spray area. One lid from each pair was removed and the soil exposed to spraying. After spraying, soils were incubated in the dark at ambient temperatures and nitrate accumulation measured at intervals for 50 days. DDT residues were measured in the samples at the end of the experiment.

There was no evidence nitrification rates were affected by spraying (Figure 2.6). About 60 % conversion of ammonium to nitrate occurred within 27 days of exposure in both sprayed and unsprayed samples. Fifty days after spraying, sprayed soils contained mean residues of 8.1 ug EDDT /g dry weight of soil and unsprayed soils 0.1 ug S DDT / g dry weight of soil. Residue levels were probably at least 25 % higher immediately after treatment.


Plant litter is shed at the time ground-spraying operations with DDT are in progress. The small number of droplets missing targeted trees were rarely carried further than 5-10 m and the resulting distribution of DDT on litter and soil was, in general, contagious. After spraying, concentrations of S DDT in soil around trees were highly variable, averaging between 1-10 ppm; background levels were 0.01-0.03 ppm.

In the dry season, risks to decomposer organisms from DDT contamination were slight as microclimatic conditions (low humidity, high temperatures and radiation) on the woodland floor presented detritivores and litter harvesters with an inhospitable environment. The small losses from litter bags placed on the surface support this thesis. Losses may have been largely due to nutrient wash out by early rains.

Litter bags buried for six months in silty loam soil demonstrated that subterranean decomposers had been active in the dry season or during some unseasonal rain. Life underground was protected from direct contamination but indirect contact with DDT with the burial of contaminated litter and via percolation of rain water was assured. Decomposition by microbes (64 um mesh) and microbes plus mites and nematodes (600 um) was unaffected by DDT, but the macro-invertebrate fauna, which processed as much again, were significantly affected. As neither baiting for termite activity nor pitfall trapping revealed consistent differences in the relative abundance of surfaceactive invertebrates in sprayed and unsprayed areas (Chapter 3) it was concluded that subterranean macro-invertebrates were being affected. However, the impact of DDT on macro-invertebrates in silty loam soil was absent in the wet season, and as 90% of buried litter was processed within 6 months, it was concluded that decomposition overall was not impaired by spraying.

The effect of DDT on macro-invertebrates was exacerbated in the sandier soils. Quantitative data on macro-invertebrate populations were not collected, but it was clear that litter processing in Julbernardia woodland in the vicinity of sprayed trees had been seriously impaired by 18-51 ppm S DDT. Edwards suggested that the greatest potential hazard of DDT residues was to the agents of litter processing in woodlands.

As the overriding factor governing the physical, biological and biochemical breakdown of organic matter is moisture, soil microbial and invertebrate activity and the related processes of soil respiration and nitrogen transformations are concentrated in the rainy season. DDT is persistent and although its half-life on tree bark and soil is short by temperate standards (50 and 80 days), residues on mopane trunks are still considerable (viz. 20 ug S DDT/cm2 bark) when the rains begin. Run-off from trees during the wet season increases soil DDT burdens nearby at a time when microbial activity, nutrient turnover and nitrogen requirements by young plants are at their highest. Quantification of the increase is difficult to estimate but wet season soil samples showed up to 500 ug EDDT/g dry weight soil (Figure 2.3).

In situ wet season respiration measurements made in the vicinity of sprayed and unsprayed trees failed to show differences attributable to DDT, despite soils containing up to 40 ug S DDT/g dry weight soil. The natural variability within sites (CV <50 %) was not high enough to mask any serious suppression of heterotrophic activity, which has only been seen at very high dose rates e.g. 50 kg/ha5. Whether the CO2 efflux truely reflected soil microbial activity is questionable with in situ measurements. However, ex situ monitoring in amended soils also provided no evidence of DDT disturbance. In all soils, the dynamics of decomposition of native organic matter and grass substrates were similar, suggesting that residues of DDT did not affect the overall microbial activity. Domsch et al. argued that duration of pesticide stress on microbial activity is more important than magnitude, and speculated that depression of soil microbial activity was only detrimental when it exceeded 60 days.

Unlike many nitrogen transformations, nitrification is often suppressed by pesticides.Residue levels of DDT in soils which were used in the nitrification study were typical of levels found around sprayed trees but fell short of concentrations shown to affect nitrification in temperate soils. That no effect of DDT on nitrification was found suggests that assimilation of nitrate by plant communities was not impaired by ground-spraying for tsetse control.


Important soil microbial activities, including litter degradation, carbon and nitrogen mineralization, were unaffected by DDT. The activities of subterranean macroinvertebrates were temporarily affected by DDT in silty clay soils, but the overall function of the decomposer ecosystem appeared unimpaired. In some sandy loam soils, however, high levels of contamination caused significant medium-term reductions in litter degradation. Given that high residue levels are extremely localised the risk of a general reduction in soil fertility in Julbernardia woodland appears slight.


1. Edwards, C.A. (1973) Persistent Pesticide in the Environment. 2nd edn. Cleveland, Ohio: CRC Press.

2. Boyer, M.G. and Perry, E. (1973) Diversity in soil fungi as affected by DDT. Mycopathologica et Mycologia applicata, 49: 255-262.

3. Cook, A.G., Critchley, B.R., Critchley, U., Perfect, T.J. and Yeadon, R. (1980) Effects of cultivation and DDT on earthworm activity in a forest soil in the sub-humid tropics Journal of Applied Ecology, 21: 21-29.

4. Perfect, T.J., Cook, A.G., Critchley, B.R. and Russell-Smith, A. (1981) The effect of crop protection with DDT on the microarthropod population of a cultivated forest soil in the sub-humid tropics. Pedobiologia, 21: 7-18.

5. Tate, K.R. (1974) Influence of four pesticide formulations on microbial processes in a New Zealand pasture soil. I. Respiratory activity. New Zealand Journal of Agricultural Research, 17: 1-7.

6. Grant, I.F. (1988) Appropriate technology for monitoring environmental effects of insecticides in the tropics. pp. 147-156, In: Field Methods for the Study of Environmental Effects of Pesticides. (Greaves, M.P., Smith, B.D. and Greig-Smith, P.W., eds) British Crop Protection Council Monograph No. 40.

7. Swift, M.J., Heal, O.W. and Anderson, J.M. (1979) Decomposition in Terrestrial Ecosystems. Oxford, UK: Blackwell Scientific Publications.

8. Swift, M.J. Russell-Smith, A. and Perfect T.J. (1981) Decomposition and mineral-nutrient dynamics of plant litter in a regenerating bush-fallow in sub-humid tropical Nigeria. Journal of Ecology, 69: 981-995.

9. Domsch, K.H., Jagnow, G. and Anderson, T. (1983) An ecological concept for the assessment of sideeffects of agrochemicals on soil organisms. Residue Reviews, 86: 65-105.

10. Atlas, R.M., Pramer, D. and Bartha, R. (1978) Asessment of pesticide effects on non-target soil organisms. Soil Biology and Biochemistry, 10: 231-239.

11. Anderson, J.R. (1978) Pesticide effects on non-target soil microorganisms. pp. 313-533. In: Pesticide Microbiology. (Hill, I.R. and Wright, S.J.L., eds) London: Academic Press.

12. Ross, D.J. (1974) Influence of four pesticide formulations on microbial processes in a New Zealand pasture soil. II. Nitrogen mineralisation. New Zealand Journal of Agricultural Research, 17: 9-17.

13. Wiese, J.H. (1964) Some biological studies on the inactivation of insecticides by various soil types. South African Journal of Agricultural Research, 7: 823-836.

14. Wiese, I.H and Basson, C.J. (1966) The degradation of some persistent chlorinated hydrocarbon insecticides applied to different soil types. South African Journal of Agricultural Research, 9: 945-970.