Different fertilization strategies can be adopted to optimize the productive components of an integrated crop–livestock systems. The current research evaluated how the application of P and K to soybean (Glycine max (L.) Merr.) or Urochloa brizantha (Hochst. ex A. Rich.) R. D. Webster cv. BRS Piatã associated with nitrogen or without nitrogen in the pasture phase affects the accumulation and chemical composition of forage and animal productivity. The treatments were distributed in randomized blocks with three replications. Four fertilization strategies were tested: (1) conventional fertilization with P and K in the crop phase (CF–N); (2) conventional fertilization with nitrogen in the pasture phase (CF + N); (3) system fertilization with P and K in the pasture phase (SF–N); (4) system fertilization with nitrogen in the pasture phase (SF + N). System fertilization increased forage accumulation from 15 710 to 20 920 kg DM ha/year compared to conventional without nitrogen. Stocking rate (3.1 vs. 2.8 AU/ha; SEM = 0.12) and gain per area (458 vs. 413 kg BW/ha; SEM = 27.9) were higher in the SF–N than CF–N, although the average daily gain was lower (0.754 vs. 0.792 kg LW/day; SEM = 0.071). N application in the pasture phase, both, conventional and system fertilization resulted in higher crude protein, stocking rate and gain per area. Applying nitrogen and relocate P and K from crop to pasture phase increase animal productivity and improve forage chemical composition in integrated crop–livestock system.

The African armyworm, the larva of the night flying moth Spodoptera exempta (Walk.), feeds on young stages of cereal crops, sugar-cane and all types of grasses in some 30 countries of Africa, from Mauritania and Senegal in the north-west to Ethiopia and Somalia in the north-east, and southwards to Namibia and South Africa. Between 1960 and 1982, outbreaks of armyworms were recorded regularly in East Africa during the ‘outbreak season’ between November/December and June/July, often from southern Tanzania towards the north or north-west at monthly intervals. Analysis of occurrence of outbreaks in Tanzania, Kenya and Uganda in the 1971/72–1973/74 seasons, and rainfall records in Tanzania for 1969–1975, showed that the occurrence of armyworms in Tanzania coincided with the onset and northward progression of the rains up to Kenya and Uganda. Additional results from monthly maps, tables, graphs and histograms of armyworm occurrences in administrative units (Districts and Provinces) in Tanzania, Kenya and Uganda for 20 years (1960–1979), showed that the onset, progress and duration of armyworm attacks in all districts and provinces in each country, tend to follow certain geographical patterns which can be monitored through the use of insect traps and Agricultural Extension Services. In this way, coordinated surveillance of priority areas could enable farmers, pastoralists, Plant Protection and Pest Control Units of the Ministries of Agriculture to identify likely areas and periods of armyworm attacks, so that they could organize strategic or emergency control of armyworms with insecticides when the larvae are still young.

Occurrences of outbreaks of the African armyworm, Spodoptera exempta, between January and December for 19 years (1961–1979), and records on nightly moth catches from light trap stations located between southern Kenya and east-central Tanzania (1–8°S and 35–41°E) for 16 years (1963–1978), have been mapped, tabulated, and analysed on a standard weekly time-table, in order to compare events occurring between the ‘off-season’ months of July to November and the beginning of the ‘outbreak seasons’ from December or January. Circumstantial evidence from the above, coupled with analysis of climatological statistics for rainfall, temperature and windflow over the area and period, support the hypothesis that populations of S. exempta can survive between southern Kenya and eastcentral Tanzania during the off-season months, and contribute to the initial development of ‘outbreak seasons’ in Tanzania or Kenya. Methods for monitoring and analysing such developments for improving the efficiency of the Armyworm Forecasting Service in East Africa are suggested.