Excessive nitrogen (N) fertilizer application and poor timing of N fertilizer application to winter wheat are common problems on the North China Plain. To study the possibilities of optimizing the timing and rate of N application, a field experiment was conducted from 1999 to 2001 in a suburb of Beijing. A control (no nitrogen) and two N fertilization strategies (conventional N application and optimized N fertilization) were designed to compare their effects on wheat growth, N nutrient status, grain yield and N balance. The conventional N fertilization strategy was given a fixed N rate of 300 kg N ha À1 , which was split, half in autumn and half in spring as a top-dressing. The timing and rate of N and application of the optimized N fertilization strategy were determined by the target value of soil mineral nitrogen demand for three growth periods of wheat, which is related to the target yield, and soil mineral N (N min ) in the effective rooting depth at the beginning of these three periods. Based on the optimized N fertilization strategy, a total of 55 and 65 kg N ha À1 had to be applied to winter wheat in the re-greening and shooting stages of the first and second experimental years, respectively. Compared with the high N rate before sowing in the conventional N fertilization treatment, the optimized N fertilization treatment did not require any N fertilizer before sowing of wheat. Despite a much lower N fertilization rate, no significant difference in N nutrient status, growth during the wheat growing period or grain yield was observed between optimized N and conventional N fertilization treatments. As a consequence of optimizing the rate and timing of the N fertilizer application to match wheat demand, a much lower residual N min and calculated apparent N loss was found as compared to the conventional N treatment. N recovery for the optimized N fertilization treatment (67% in 1999/2000 and 66% in 2000/2001) was much higher than that of the conventional N fertilization treatment (19% in 1999/2000 and 18% in 2000/2001). In conclusion, the optimized N fertilization strategy can synchronize N demand of wheat and the N supply from soil and fertilizer, and therefore drastically reduce N application rates without any yield losses.

A survey on current fertilizer practices and their effects on soil fertility and soil salinity was conducted from 1996 to 2000 in Beijing Province, a major vegetable production area in the North China Plain. Inputs of the major nutrients ͑NPK͒ and fertilizer application methods and sources for different vegetable species and field conditions were evaluated. Excessive N and P fertilizer application, often up to about 5 times the crop requirement in the case of N, was very common, especially for high-value crops. Potassium supply may have been inadequate for some crops such as leafy vegetables. Urea, diammonium orthophosphate ͑͑NH 4 ͒ 2 HPO 4 ͒ and chicken manure were the major nutrient sources for vegetable production in the region. Over 50% of N, 60% of P and nearly 90% of K applied originated from organic manure. Total N application rate for open-field Chinese cabbage from organic manure and inorganic fertilizers ranged from 300 to 900 kg N ha -1 on 78% of the farms surveyed. More than 35% of the surveyed greenhouse-grown tomato crops received Ͼ 1000 kg N ha -1 from organic and inorganic sources. A negative K balance ͑applied K minus K removed by the crop͒ was found in two-thirds of the surveyed fields of open-field Chinese cabbage and half of the surveyed fields of greenhousegrown tomato. Plant-available N, P and K increased with increasing length of the period the greenhouse soils had been used for vegetable production. Similarly, soil salinity increased more in greenhouse soils than in open-field soils. The results indicate that balanced NPK fertilizer use and maintenance of soil quality are important for the development of sustainable vegetable production systems in this region.

In den Jahren 1988 und 1989 wurden auf verschiedenen Standorten und bei unterschiedlichen Umweltbedingungen mit Hilfe der Integrated Horizontal Flux Methode Messungen der NH3‐Verluste nach Ausbringung von Rinderflüssigmist durchgeführt. Ziel dieser Untersuchungen war es, die NH3‐Verluste zu quantifizieren sowie die dafür verantwortlichen Einflußfaktoren zu erfassen und zu bewerten. Je nach Umweltbedingungen und Infiltration betrugen die NH3‐Verluste der in dieser Arbeit dargestellten Versuche zwischen 12 und 65% der ausgebrachten NH4‐N‐Menge. Aus der Vielzahl der Einflußfaktoren wurden drei leicht zu messende bzw. zu schätzende Faktoren ausgewählt. Mit Hilfe dieser Faktoren wurde ein Schema zur Abschätzung von NH3‐Verlusten entwickelt. In diesem Schätzrahmen werden in Abhängigkeit von Infiltration, durchschnittlicher Tagestemperatur, Niederschlagshöhe sowie Stunden bzw. Tagen (mit oder ohne Einarbeitung) nach Ausbringung die zu erwartenden NH3‐Verluste ermittelt. Der Rahmen soil eine praktische Hilfe zur Abschätzung der NH3‐Verluste in Bezug auf die Menge des ausgebrachten NH4‐Stickstoffs geben und die Möglichkeiten der Verminderung dieser Verluste in Abhängigkeit von Umweltbedingungen und Zeitpunkt der Einarbeitung aufzeigen.

A windtunnel system is presented applicable for measuring ammonia emissions under field conditions. With this system two objectives are achieved: No alteration of the microclimatic conditions in the testing area Reliable determination of the volumetric air flow and the ammonia concentration The use of a transparent foil, the precise adjustment of the flow velocity to the windspeed outside, and the constant cross‐sectional area over the whole length of the tunnel are the most important constructional details of the system.

Advanced nitrogen (N) advisory systems require target values of N supply for the crop and the results of soil testing for inorganic nitrogen (N min ) before cultivation and estimates of N release from mineralization. Two field experiments with different N supply levels were conducted in the Beijing region to determine the target values of N supply and N mineralization rates for optimization of N fertilization of Chinese cabbage (Brassica campestris L. ssp. Pekinensis) and carrot (Daucus carota L.). Crop yield, N uptake, and ORDER REPRINTS soil inorganic N were investigated during the crop growth periods. Marketable yields of the two crops increased significantly with N application rate, and were described as linear-plateau functions (Chinese cabbage: Y ¼ 0.16N þ 80.5, when 0 < N 319; Y ¼ 112.7, when N ! 319; and carrot: Y ¼ 0.12N þ 50.5, when 0 < N 247; when Y ¼ 67.1, N ! 247; where: Y was marketable yield, t FW ha À1 ; N was N supply level, the sum of soil inorganic N at preplanting and applied fertilizer N, kg N ha À1 ). Similar trends were also observed in crop N uptake. The agronomically effective N supply levels for Chinese cabbage and carrot growth in Beijing region were 349 and 277 kg N ha À1 , respectively, for target marketable yields of 120 t FW ha À1 for Chinese cabbage and 65 t FW ha -1 for carrot considering extra 30 kg N ha À1 of fertilizer to avoid the risk of crop yield reduction. During the crop growth period net N mineralization in the root zone was 38 kg N ha À1 for Chinese cabbage and 46 kg N ha À1 for carrot. Weekly N mineralization rate was about 3.2 kg N ha À1 . Nitrogen loss coefficient, defined as the ratio of average weekly apparent N loss to N supply level, ranged from À0.011 to 0.053 for the different N supply levels in the two experiments. The N net mineralization rates determined, N loss rates and target values were found to be useful for prediction of optimum N supply based on N balance calculations for sustainable vegetable production.

A series of field experiments were conducted over a three-year period to test a modified N-Expert system (the decision support system for nitrogen recommendations) with different irrigation regimes for a rotation of amaranth (Amaranthus tricolor L.), spinach (Spinacia oleracea L.), and cauliflower (Brassica oleracea L.). Local commercial fertilizer practice based on farmer surveys was selected as the control, and irrigation treatments followed conventional practice with a balanced irrigation schedule. Some parameters, notably target yields and net mineralization rates, were modified according to local conditions, and most were referenced from the database of the N-Expert system in the first cultivation year of the rotation. Parameters were tested in earlier field experiments and were fine-tuned in later experiments. No significant yield reduction occurred below conventional nitrogen (N) practice using the recommended N treatment (N2) with different water treatments, except for irrigation treatments between 60% and 90% of PESW (plant extractable soil water: the difference between field capacity and wilting point) in 2000 and conventional irrigation practice in 2001 for cauliflower. There was a significant decrease in residual N min at harvest in treatments using the modified N-Expert recommendation system compared with conventional N practice. This indicates that there is considerable potential for use of the N recommendation system for sustainable vegetable production in the North China Plain.

Field experiments with silage maize were conducted in 1987 and 1988 on a loess‐derived Luvisol in southwest Germany. Four nitrogen fertilizer treatments were compared: application of preplanting NH4 N (plus a nitrification inhibitor, dicyandiamide as Didin) and preplanting NO3‐N, split application of NO3‐N (preplanting and side dressed 45 days after planting) and a control without nitrogen fertilizer in 1987 and with 64 kg N ha−1 as calcium ammonium nitrate in 1988. The total amounts of soil mineral nitrogen (Nmin+ fertilizer N) were 200 kg N ha−1 in 1987 and 240 kg N ha−1 in 1988. Suction cups and tensiometer were installed at five depths and samples were taken in regular intervals. Nitrate concentrations in the suction solution steeply increased at 15 cm and 45 cm soil depth 3‐4 weeks after fertilizer application (1987 up to 160mgNl−1; 1988 up to 170mgN l−1) and steeply decreased up to 75 cm depth with the onset of intensive N uptake at shooting. Ammonium concentrations in the suction solution were very low (0‐0.16 mg N l−1). Compared to preplanting NCyN application, preplanting NH4‐N and split NO3‐N application decreased nitrate concentrations in the suction solution in spring 1987. In 1988, however, nitrate concentrations in the suction solution of preplanting NH4‐N and split NO3‐N application plots did not fall below 50mgNl−1 at 15 cm depth during the growing season. Nitrate concentrations of split NO3‐N application increased again in autumn 1988 and hence doubled the calculated N losses by leaching during the winter months compared to preplanting N applications. At shooting, plants of the preplanting NH4‐N treatment had lower nitrate concentrations in leaf sheaths compared to plants of preplanting NO3‐N application. Total N uptake of maize between shooting and early grain filling of preplanting NH4‐N and split NO3 ‐N application tended to be higher compared to preplanting NO3‐N application, reflecting the higher N availability in the soil later in the season. However, final dry matter yields and N uptake were not significantly affected by N form or time of N application. Since N losses by nitrate leaching between N application and onset of N uptake by plants were negligible on the experimental site, preplanting NH4‐N application and split NO3‐N application showed no agronomic advantages. High amounts of side dressed NO3‐N may increase nitrate leaching during the winter months, especially in years with delayed rainfall after application.

No abstract