«: AGROCHEMICALS: FATE IN FOOD AND THE ENVIRONMENT PROCEEDINGS OF A SYMPOSIUM, ROME, 7 - 1 1 JUNE 1982 JOINTLY ORGANIZED BY IAEA AND FAO l^J I N T E R ...»
Over four years Sotiriou and Korte [ 3 6 ] evaluated the balance and fate o f urea ( C O ( 1 5 N H 2 ) 2 ) in sandy and loamy soils, with and without addition o f organic matter, and in loamy soil with different crops. They found that the applied nitrogen decreased linearly with the soil depth for sandy soil and exponentially for loamy soil. Organic matter additions did not affect losses. Youngdahl et al. , in a study to evaluate the potential losses o f urea o f different granule size in soils o f different textures, showed that for moderate and high drainage rates (say 10 — 20 mm), especially with soils of low CEC, the leaching losses of supergranulated urea can be total. In sandy loam 15 N-analysis showed that the crop (rice) recovered 63% of the supergranulated fertilizer when the drainage rate was
4.4 mm/d but recovered only 5% for a 18.3 mm/d rate. The leaching losses o f supergranulated urea were always lower than the losses for other nitrogen fertilizers. In Sri Lanka Golden  reports that normally 200 kg N/ha are applied to tea crops and that the efficiency o f utilization is in the range o f 30 to 50%, the remainder being potentially available for leaching.
Libardi and Reichardt , studying the fate o f urea- 15 N (120 kg N/ha) applied to a bean (Phaseolus vulgaris L.) c r o p o n an OxicPaleudalf in Brazil, found that during the cropping season ( 1 2 0 d) 6.7 kg N/ha were lost by leaching, this including both fertilizer and soil nitrogen. The contribution o f fertilizer nitrogen to leaching could not be estimated owing to the small size o f the samples obtained from porous cup extractors (Reichardt et al. [ 3 9 ] ). In these studies they were faced with the problem o f spatial variability o f soil properties, which was found to be very significant. Reichardt et al. [40, 41 ] found that for this Oxic Paleudalf the spatial variability of soil bulk density, soil water retention curves and soil hydraulic conductivities could not be neglected. Nascimento et al.  and Reichardt et al. , using a pulse o f 36C1 during steady infiltration o f water into the same soil, found saturated hydraulic conductivities varying from 1.26 to
6.86 cm/h inside a 10 X 10 m plot. Using Gaussian statistics they obtained an average value o f 1.513 ± 0.687 cm/h. The time for 36C1 to reach to a depth o f 120 cm varied from 38.5 to 95 h.
Meirelles et al. [ 4 4 ] and Cervellini et al. , in the same soil , applied 100 kg N/ha as ammonium sulphate- 15 N to a bean crop (Phaseolus vulgaris L.) and found that a total o f 15 kg N/ha were leached with only 1.35 kg o f this derived from fertilizer. These are average values for which spatial variability was not taken into account.
Reichardt et al. , also in Brazil, studied the dynamics o f nitrogen (80 kg N/ha, ( 1 S N H 4 ) 2 S 0 4 ) applied to a maize crop (Zea mays L.) on sandy IAEA-SM-263/36 Oxisol, for which spatial variability was a minor problem . They found that
9.2 kg N/ha were lost below the 127.5 cm depth by leaching, with only 4.8% derived from fertilizer, which gives a total o f 0.4 kg N/ha. The study covered a period o f 150 d during the rainy season, but rainfall was very well distributed and as a result the crop recovered 89.1% o f the applied fertilizer nitrogen.
Libardi et al. , in order to increase understanding o f the fate o f applied nitrogen in a Phaseolus vulgaris crop grown under tropical conditions, applied 15 N-labelled urea to bean crops grown on Oxic Paleudalf  and followed with a pulse o f enriched 15 N-fertilizer for three consecutive cropping periods. Results showed that 31.2% o f the nitrogen in the first crop was derived from the applied urea (100 kg N/ha), which represents a nitrogen utilization efficiency o f 38.5%.
The second crop had 6.2% o f the fertilizer nitrogen applied to the first crop, and only 1.4% o f the nitrogen in the third crop was derived from fertilizer. In total, the three crops recovered 44.3% o f the nitrogen applied to the first crop, and the remainder was either still in the soil profile or had been lost by leaching, volatilization or denitrification. Nitrogen-15-enrichment o f mineral nitrogen ( N 0 3 + N H 4 ) suggests that at the end o f the second crop the pulse o f fertilizer applied to the first crop had probably passed the 120 cm depth. Nitrogen-15enrichment o f organic nitrogen suggests that the root activity o f beans and weeds transported nitrogen to the 9 0 — 1 2 0 cm depth (or deeper). They accounted for 109 kg fertilizer nitrogen per hectare in harvested biomass, crop residue and soil at the end of the first cropping period. This indicates an experimental error of about 10%; on nitrogen it was lost by volatilization, denitrification or leaching below 120 cm. 15 N-data at the end o f the first cropping period support the assumption o f leaching. At the end o f the second and third cropping periods 76 and 80 kg N/ha, respectively, could be accounted for, suggesting that 20 to 25% o f the applied nitrogen was lost over a two-crop period.
Calvache  and Araújo Silva  studied in detail the fate o f fertilizer nitrogen in a corn (Zea mays L.) crop on Oxic Paleudalf  in Brazil. During the cropping period 30% (195 mm) o f the water balance corresponded to drainage below the 120 cm depth. Assuming that nitrates are leached only by mass flow, they accounted for a loss o f 32.4 kg N/ha, with a contribution o f 34% from fertilizer ( 11.0 kg N/ ha). The application rate was 100 kg urea-N/ha.
In the latest study in Brazil, Urquiaga  used a highly enriched (56% 15N atom excess) ammonium sulphate (42 kg N/ha) on a bean crop grown on Oxic Paleudalf . The crop had an extremely high fertilizer efficiency use o f 76% (32.0 kg N/ha). At the end o f the cropping period 9.2 kg N/ha derived from fertilizer were found in the soil profile. Leaching losses were not calculated, but even assuming no other losses, leaching would amount to only 0.8 kg N/ha.
Table I summarizes the nitrogen leaching data found by seven authors. Using average values, which certainly has many limitations, it can be seen that for nitrogen application rates o f about 90 kg N/ha, under tropical conditions, only 284 REICHARDT et al.
4.5 g o f fertilizer nitrogen are lost b y leaching, per hectare and per millimetre o f rainfall. It is clear that a minor p r o p o r t i o n o f the leached nitrogen derives f r o m fertilizer, and it can be concluded that losses b y leaching for fertilizer applied at normal levels are not a problem.
3.2. Run-Off Losses The soil surface layer is generally the richest in nutrients and therefore its loss can, in many cases, bring decreases in fertility and productivity levels. In tropical areas water erosion is the most important, and is related t o conditions o f climate, soil, relief, c r o p cover and management practices. So, due t o the different factors that affect r u n - o f f and erosion, information about losses is also very variable. On the other hand, most reports deal with soil losses as a whole, generally expressed as tonnes o f soil per hectare per year, and little attention is given t o fertilizer losses.
Lai et al. [50, 51 ] report that in Nigeria large amounts o f soil and nutrients are lost b y run-off and erosion when Alfisols with a sandy surface are left without crop protection. Under these conditions soil losses can amount t o 115 t/ha per year, productivity is reduced t o less than 50% and additional fertilization does not IAEA-SM-263/36 285 correct the fertility state o f the soil. This, however, is an extremely high loss.
In Brazil Bertoni et al.  report losses ranging from 0.9 to 27 t/ha per year, and most o f the values are less than 15 t/ha per year. A loss o f 15 t/ha per year represents a loss o f soil layer o f 1.2 mm thickness, considering a soil bulk density o f 1.3 g / c m 3.
Therefore, it seems that erosion is not the main cause o f the low fertility o f tropical soils. Nye and Greenland [ 1 9 ] report that on a cultivated forest soil the greatest nutrient concentration is in the first 7.5 cm, even though erosion losses are minimal when the slopes are not greater than 10%. Owing to the mobility o f ions like N 0 3, SO4, СГ, they are generally leached downwards before being lost by run-off.
3.3. Gaseous losses
Volatilization o f ammonia occurs with high probability when fertilizer is applied to the soil surface, but it depends on the nature o f the fertilizer, soil CEC, pH, organic matter content, and other properties like temperature and water content. In the case o f urea, hydrolysis forms ammonium carbonate in the presence o f urease, an enzyme o f high activity in soils o f high organic matter content and high CEC. The carbonate, owing to its low stability, breaks down into ammonia and carbon dioxide. Gasser  reports that in slightly acid or neutral soils, hydrolysis o f urea can increase soil pH and stimulate N O j accumulation, and thereby increase the probability o f formation o f volatile compounds ( N 0 2, NO), which are subject to losses. This mechanism seems to operate more efficiently with urea than with ammonium salts. On the other hand, Gasser stresses that the pH increase due to hydrolysis in acid soils favours the nitrification o f ammonium, it being faster for ammonium derived from urea than from other salts.
NH 3 losses are c o m m o n in soils with a high calcium content or in those that have received heavy liming, but quantities vary mainly with fertilizer type.
According to Malavolta  and Terman , in ammoniacal fertilizers the complementary anion has a great influence on NH 3 losses; the following sequence is considered to be valid for the losses: urea ammonium sulphate ammonium nitrate monoammonium phosphate.
In Brazil Anjos and Tedesco  studied the volatilization o f ammonium derived from urea and ammonium sulphate. For urea losses ranged from 12 to 30% and for ammonium sulphate only from 0.5 to 1.1%. Libardi and Reichardt , in their study o f the fate o f fertilizer nitrogen, did not find volatilization losses for urea applied in the furrow o f an Alfisol, probably due to the soil pH which was 5.5.
In summary, according to Terman , N-NH 3 losses are practically eliminated in acid soils when nitrogen fertilizers are applied at depths greater than 286 REICHARDT et al.
5 cm. Surface applications have a high risk o f loss, especially in soils o f low CEC, high pH and low water content.
Losses in the form o f N 2 0 and N 2, that is by denitrification, are more significant under reducing conditions, which occur in soil profiles with deficient drainage. Terman [ 5 4 ] reports that the low recovery o f applied nitrogen in flooded rice crops is usually due to N 2 0 and N 2 losses to the atmosphere. Also, Thenabadu  states that gaseous losses o f nitrogen are o f special significance in flooded rice growing in soils where ammonium salts could be nitrified on the soil surface and which, on reaching the lower reduced horizons, get denitrified.
Many reports are available and in many cases much o f the nitrogen that could not be accounted for in attempts to prepare balance sheets is considered lost to the atmosphere in gaseous forms.
Another important aspect is nitrogen losses from the tops o f plants, which are discussed in detail by Wetselaar and Farquhar . They report that for annual field crops the losses reach 40 — 50 kg N/ha, at mean rates o f 1.2 kg N/ha per day, between maturity and harvesting.
3.4. Crop extraction and export
It is generally assumed that crop extraction is the total quantity of a given element accumulated in the whole plant, and that crop export is the fraction o f the total within the harvested product. In most cases the non-exported material remains on the field as harvest residue.
As was mentioned in Section 2.1 crop needs are strongly related to plant species and are affected by factors such as climate and the technology level, which certainly affect yield (or crop export). The available literature on the subject is enormous and a complete survey is not appropriate here. An analysis o f data presented by Malavolta  and Sanchez  gives a good idea o f the order o f magnitude o f crop exports. They report that under tropical conditions export o f nutrients by the main crops (corn, rice, wheat and sorghum) range from 20 to 25 kg N/ha per tonne o f product, corn and sorghum being the best 'exporters' o f nitrogen, due to their high productivity. In tubers and roots, nitrogen exports range between 3.8 and 5.4 kg/t per hectare. Grain legumes show much higher values (32 to 40 kg N/t per hectare) due to their high protein content. Sugar-cane crops export practically all the extracted nitrogen, accounting for 0.8 to 1.3 kg N/t per hectare.
[1 ] MALAVOLTA, E., "Manual de química agrícola, Adubos e adubaçâo", 3rd éd., Ceres, Sâo Paulo (1981) 31.
 BREMNER, J.M., "Total nitrogen", Methods of Soü Analysis, Part 2, (BLACK, C.A. et al., Eds), American Society of Agronomy, Madison (1965) 1149.
 CHENG, B.T., "Soil organic matter as a plant nutrient", Soil Organic Matter Studies (Proc. Symp. Braunschweig, 1976), IAEA, Vienna (1977) 31.
 FASSBENDER, H., Química de suelos con énfasis en suelos de América Latina, IICA, Turrialba (1975).
 VAN RAIJ, В., Avaliaçâo da fertilidade do solo, Potash and Phosphate Institute and International Potash Institute Piracicaba, Brazil (1981) 142.
 GREENLAND, D.J., Nitrate fluctuations in tropical soils, J. Agrie. Sci. Camb. 50 (1958) 82.
 LEAL, J.R., ALVAHYDO, R., Transformaçâo e deslocamento do ion amonio em solo da Série Itaguaí, Pesq. Agropec. Bras., Sér. Agron. 6 (1961) 129.
 SANCHEZ, P.A., Properties and Management of Soils in the Tropics, John Wiley and Sons, New York (1976) 184.
 VAN WAMBEKE, A., "Propriedades que influyen en el manejo de los oxisoles en ecosistemas de sabana", Manejo de suelos en la América tropical (Proc. Symp. Cali, 1974), (BORNEMISZA, E., ALVARADO, A., Eds), North Carolina State University, Raleigh, (1975) 371.
 VERDADE, F.C., Estudo da variabilidade dos nitratos num solo tipo Terra Roxa Misturada, Bragantina 11 (1951) 269.
 COOKE, G.W., "The fate of fertilizers", The Chemistry of Soil Processes (GREENLAND, D.J., HAYES, M.H.B., Eds), John Wiley and Sons, New York (1981) 563.
 EPSTEIN, E., Mineral Nutrition of Plants, Principles and Perspectives, John Wiley and Sons, New York (1972) 286.