Agricultural and Food Science 166 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 166–182. © Agricultural and Food Science Manuscript received March 2005 Weed occurrence in Finnish coastal regions:   a survey of organically cropped spring cereals Paul Riesinger Department of Applied Biology, FI-00014 University of Helsinki, Finland, present address: Edenkatu 5, FI-10300 Karjaa, Finland, e-mail: paul.riesinger@sydvast.fi Terho Hyvönen MTT Agrifood Research Finland, Plant Production Research, FI-31600 Jokioinen, Finland Weed communities of organically cropped spring cereal stands in the southern and the northwestern coast- al regions of Finland (= south and northwest, respectively) were compared with respect to number of spe- cies, frequency of occurrence, density and dry weight. Regional specialization of agricultural production along with differences in climate and soil properties were expected to generate differences in weed com- munities between south and northwest. Total and average numbers of species were higher in the south than in the northwest (33 vs. 26 and 15.6 vs. 10.0, respectively). Some rare species (e.g. Papaver dubium) were found in the south. Fumaria officinalis and Lamium spp. were found only in the south. The densities and dry weights of Lapsana communis, Myosotis arvensis, Polygonum aviculare, Tripleurospermum inodorum and Vicia spp. were higher in the south, while the densities and dry weights of Elymus repens, Persicaria spp. and Spergula arvensis were higher in the northwest. Total density of weeds did not differ between south and northwest (average = 565 vs. 570 shoots m-2, respectively). Total dry weight of weeds was higher in the northwest compared with the south (average = 1594 vs. 697 kg ha-1, respectively), mainly due to the high dry weight of E. repens. The only variable that was dependent on the duration of organic farming was weed density in the south. The abundance of nitrophilous in relation to non-nitrophilous weed species was higher while the abundance of perennial ruderal and grassland weed species was lower compared with previous weed surveys. This can be regarded as the result of increasing cropping intensity on organic farms in Finland. Different weed communities call for the application of specific target-oriented weed manage- ment in the respective coastal regions. Key words: cereals, coastal regions, farm case studies, Finland, organic farming, weed flora, weed manage- ment, weed occurrence 167 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 166–182. Introduction Weed community composition and weed abun- dance are influenced by climate and soil proper- ties, thus leading to regional differences. The de- velopment of regional differences in arable weed communities is further affected by specialization of agricultural production, changes in production intensity and conversion to alternative farming concepts (El Titi 1995, Mäder et al. 2002). In Fin- land, this development has been studied by recur- rent weed surveys. In the first national weed survey carried out in Finland, Mukula et al. (1969) attributed geographi- cal differences of frequency of occurrence and abundance of weed species to differences in climate and soil properties. Since the 1960s, mechanization and specialization have brought with them a de- cline in the number of farm enterprises in Finland. As a result of regional specialization, stockless crop husbandry has become predominant in southern Finland, while animal husbandry prevails in cen- tral, eastern and northwestern Finland. At the same time, the crop rotations have become more monoto- nous and production has been intensified (National Board of Agriculture 1991, Information Centre of the Ministry of Agriculture and Forestry 2003). Commenting on the second national weed survey, Erviö and Salonen (1987) pointed out that the weed flora composition in conventionally cropped fields was changing due to interacting factors related to the intensification of crop production. Since 1991 when a financial support program for organic agriculture was launched in Finland, the area devoted to organic agriculture has in- creased from 0.3% (1991) to 7.0% (2002) of the total area of arable land. In 2002, 6.8% of all active farms in Finland were acknowledged as organic farms (Plant Production Inspection Centre 2003). Both conventional and organic farms in Finland strive for advantages of scale, i.e. increase in size of arable land, specialization on a few branches of production and high proportions of cash crops (Plant Production Inspection Centre 2003). Monotonous crop rotations along with fertili- zation and application of herbicides have been found to be detrimental to the species diversity of weed communities. Monotonous crop rotations contribute to a decrease in weed species richness (Hald 1999, Leeson et al. 2000). However, out of all agricultural management practices, the applica- tion of herbicides exerts the largest influence on weed species diversity and abundance (Doucet et al. 1999, Hyvönen and Salonen 2001). In Finland, the application of herbicides on agricultural land started to increase after the first nation-wide weed survey was carried out 1961–1964 (Mukula et al. 1969). Another measure that has a selective impact on weed species is seed cleaning (Svensson and Wigren 1986). Intensive soil cultivation, on the other hand, leads to a more general decline of weed species diversity and abundance (Cardina et al. 1991, Frick and Thomas 1992). High soil nutrient levels favour shade-tolerant and competitive weed species (Ellenberg et al. 1991, Blackshaw et al. 2003) but suppress others and lead generally to a decrease of weed species diversity and abundance (Erviö 1972, Pyšek and Lepš 1991). Organic farming has been found to support populations of rare and threatened weed species (Rydberg and Milberg 2000) and to contribute to larger species diversity (Hald 1999, Hyvönen and Salonen 2001, Hyvönen et al. 2003). However, or- ganic farming per se does not necessarily contrib- ute to higher species diversity. In fact, intensifica- tion of weed management on organic farms has been observed to have a negative impact on weed species diversity (van Elsen 2000). A combination of intensive tillage, high soil nutrient status, high- yielding crop varieties and efficient mechanical weed control may result in organically managed crops that are highly competitive in relation to the weed flora (Albrecht and Mattheis 1998). Legume- rich leys, grain legumes and legume-cereal inter- crops provide symbiotically fixed nitrogen (Jensen 1986, Nykänen et al. 2000). In some cases large amounts of manure are transferred from conven- tional to organic farms. Weed abundance in organically cultivated crop stands has been found to be higher than in conven- tionally cropped stands (Leeson et al. 2000, Salo- nen et al. 2001a). Salonen et al. (2001b) empha- sized that the spreading of especially Chenopodi- 168 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Riesinger, P. & Hyvönen, T. Weed occurrence in coastal regions of Finland um album, Cirsium arvense, Elymus repens and Sonchus arvensis might seriously threaten the fur- ther adoption of organic farming in Finland. Lack of understanding of the specific features of an or- ganic cropping system may cause a rapid increase of weed infestation (Rantzau et al. 1990, Freyer 1991). On the other hand, Mela (1988), surveying pioneer farms of organic agriculture in Finland, found that weed biomass decreased as a function of the period of organic production (while weed den- sity increased). According to Davies et al. (1997), a rapid increase of weed density and weed seedbank numbers during the first years of conversion might be followed by a stabilization, provided that ley and intensively managed horticultural crops make up a part of the crop rotation. After all, the manage- ment skills of the farmer are decisive for the level of weed infestation (Rasmussen et al. 1998). Owing to the geographical differences in weed flora composition and abundance found in the first national weed survey carried out in Finland, Mukula et al. (1969) proposed a floristic division of Finland into different zones, among others the southwestern archipelago and the northwestern re- gion along the Gulf of Bothnia. The following two national weed surveys concentrated on the inland regions of Finland (Erviö and Salonen 1987, Salo- nen et al. 2001a). Salonen et al. (2001a) confirmed the contribution of the regional specialization of plant cropping to the differences between weed floras in the southern and southwestern regions as well as between the central and eastern regions. However, the impact of specialization and intensi- fication on the weed floras in the southern and the northwestern coastal regions remains to be investi- gated. Nothing is known about the impact of or- ganic cropping practices on the weed communities of the Finnish coastal regions since the 1980s (Mela 1988). This study aims to supplement the latest na- tional survey of weed occurrence (Salonen et al. 2001a, b) by investigating organically cropped spring cereal stands in the southern and the north- western coastal regions. Weed communities were expected to differ from each other due to different geographical location and regional specialization of agricultural production. Furthermore, it was ex- pected that the weed community composition in the coastal regions would differ from that of the inland regions. As a result of specialization and in- tensified production methods, weed communities in organically cropped spring cereal stands along the Finnish coast were expected to be less diverse and less abundant than the weed communities that had been recorded in these regions in convention- ally cropped spring cereals 1961–1964 (Mukula et al. 1969) and in organically cropped cereal stands 1984–1986 (Mela 1988). Organic farms in the coastal regions were expected to differ from each other with regard to weed abundance, depending not only on climatic and edaphic conditions but also on agricultural management. Material and methods Study area, farms and fields Weed communities in organically cropped spring cereal stands were studied by a survey that was carried out 2002. The fields were located along the Gulf of Finland in the south, in the southwestern archipelago and along the Gulf of Bothnia in the northwest. The southern coastal region includes the districts of Itä-Uusimaa, Uusimaa, Varsinais- Suomi and Ahvenanmaa (hereafter = south). The northwestern coastal region covers the district of Pohjanmaa (hereafter = northwest). The investi- gated coastal regions stretch from about 60° to 65° North and differ from each other with regard to cli- mate, soil properties and branches of production. Precipitation is lower and soil types differ in the coastal regions compared with the adjacent inland regions (Kurki 1982, Mukula and Rantanen 1987, Finnish Meteorological Institute 2003). Thirty farms managed according to the regula- tions of organic farming (European Council 1991, 1999) were selected with the help of the regional extension services. Farm and field data are speci- fied in Table 1. The large geographical area cov- ered by this study and limitations with respect to labour advocated a restricted sample size. 169 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 166–182. farms were engaged in animal husbandry. Howev- er, 10 out of the 14 stockless farms imported ani- mal manure from other farms situated in their neighbourhood. Animal husbandry farms and stockless farms were not evenly distributed along the coastal regions but reflected structural differ- ences similar to the regional distribution of con- ventional farms. In the south as well as in the southernmost part of the northwest, the farms were mainly stockless or kept beef cattle, while all farms in the northern part of the northwestern region spe- cialized in milk production. On every farm, one field cropped with spring cereals was randomly selected. Fields were not chosen to be representative of the respective farms. Instead, the objective was to establish samples of weed communities characteristic of the southern and the northwestern coastal regions, respectively. Pure stands of spring wheat, oats or barley were preferred. In five cases no pure stand of spring ce- reals was grown and a crop with undersown ley was therefore chosen. The average size of the fields was 3.2 ha (median = 2.3 ha), with a range between 0.5–11.0 ha. Soil data were derived from samples taken by the farmers and analyzed according to the routines described by Vuorinen and Mäkitie (1955). With the exception of one field, primary tillage had been carried out by ploughing. Stubble culti- vation with spring-tine cultivators had been carried out in 6 out of 30 fields prior to ploughing. Twen- ty-four fields had been ploughed during autumn while 5 fields had been ploughed in spring. Spring cereals had been sown during April on Ahvenan- maa, during the first half of May in the other south- ern districts, and during the second half of May in the northwest. Weed harrowing was applied in 5 out of 30 cereal stands. Weather conditions The growing period in the study year (calculated in days when the base temperature of 5°C was ex- ceeded) lasted 163 days in the south and 148 days in the northwest. The effective temperature sum (calculated in degree days accumulated above the Table 1. Farm and field data. Southern coastal region Northwestern coastal region Number of fields a 18 12 Branch of production animal husbandry 9 7 crop husbandry 9 5 Proportion of ley or fallow in rotation, % 10–29 5 5 30–39 2 6 40–70 11 1 Proportion of winter cereals in rotation, % 0–10 3 1 11–20 4 2 21–30 4 – >30 1 – Soil type b clay 13 3 silt 3 4 sand 2 3 organic – 2 Year of conversion c 1981–1990 1 2 1991–1995 10 5 1996–2000 7 5 Cereal crop wheat 13 4 barley 1 3 oats 3 3 mixture 1 2 a Number of fields corresponds to number of farms. b Humus contents varied between 3–5.9% in most of the clay and silt soils, between 6–11.9% in some clay, silt and sand soils and between 12–19.9% in a few clay and silt soils. The organic soils investigated in northern Pohjanmaa contain 20–39.9% humus and are classified as mull soils. c Year of conversion refers to the surveyed field. The farms participating in this survey cultivat- ed on average 67.7 ha of arable land (median = 53.3 ha) whereas the average Finnish organic farm comprised 30.9 ha of arable land (Plant Production Inspection Centre 2003). Sixteen out of the 30 170 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Riesinger, P. & Hyvönen, T. Weed occurrence in coastal regions of Finland base temperature of 5°C, DD5) reached 1536– 1693ºC in the south and 1405–1499ºC in the northwest. Precipitation during April, May, June and July varied from 117.7 mm to 275.9 mm in the different districts included in the survey. Precipita- tion was generally lower at the south coast than in the inner parts of the southern districts or in the northwest. Compared with the average data of the latest 30 years period, the growing period in 2002 was characterized by a high effective temperature sum. Precipitation was low in April and May but high in June and July (Finnish Meteorological In- stitute 2003). The investigation comprised of only one year but included different climatic zones, dif- ferent soil types and different management sys- tems. Sampling Number of weed species (species m-2), weed den- sity (shoots m-2) as well as dry weights (g m-2) of crop and weed species were assessed at the stage when both weeds and crop had reached their max- imum biomass, i.e. at DC (decimal code) 70–79 on the scale of Zadoks et al. (1974). Because the fields were situated along the coastline from the south- east to the northwest, sampling could be conducted within a narrow range of crop development stag- es. Due to the plentiful occurrence of weeds and limitations with respect to labour, sampling was restricted to four plots per field. The sampling plots were chosen by dividing the longest diagonal across the field into five even distances. Headland, shallow sites suffering from drought, compacted sites or sites close to artesian springs were avoided. Sampling was carried out with a rectangular metal frame measuring 0.25 m2. The frame was placed so that the distance from crop row to frame was the same both in the upper and lower margin of the frame. Crop and weed species were cut at soil sur- face, sorted by species, counted and dried in paper bags. The samples were dried by airflow dryer at 30–40ºC. The samples were weighed to the nearest 0.1 g. Density and dry weight were used as meas- ures of weed abundance. Assessment of weed species The plant species nomenclature follows that of Hämet-Ahti et al. (1998). The full scientific names with attributions together with the BAYER codes of weeds (Bayer 1992) are given in Table 2. Spe- cies that could not be identified were pooled by genera and taxa. Individuals belonging to the taxa Brassica spp., Chenopodium spp., Galeopsis spp., Galium spp., Lamium spp., Persicaria spp., Rumex spp., Taraxacum spp., Trifolium spp., Veronica spp. and Vicia spp. were not identified to species level. Only few individuals of the species Matri- caria matricaroides (LESS.) PORT. and Matri- caria recutita L. were found, and these were there- fore included in the species Tripleurospermum in- odorum (L.) SCH. BIP.; likewise, Viola arvensis includes V. tricolor. Grasses other than Elymus re- pens were treated as one taxon. Although ley grasses and Trifolium species in pure stands of ce- reals cannot be considered to be part of the crop, they were not regarded as weeds with respect to weed density. The competitive potential of ley spe- cies was expressed by including the biomass of ley species either into the weed biomass or, in the case of undersown ley seed, into the biomass of the nurse crop. Analyses of the data The Wilcoxon two-sample test (see Sokal and Rohlf 1995) was applied to compare average num- bers of species, densities and dry weights of crop and weeds as well as single weed species between the southern and the northwestern regions. The re- lationship between the duration of organic farming and the number of species, the density, and the dry weight of weeds was studied by linear regression in both regions separately. For this purpose, den- sity and dry weight data were log (x + 1) trans- formed. Since the number of sampled fields varied be- tween south and northwest, the total species num- bers of these regions could not be compared. The expected number of species E(Sn) for each region was therefore calculated by rarefaction. In rarefac- 171 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 166–182. tion, the number of species of larger samples is scaled down to a given number of individuals that permits the comparison of the numbers of species between samples differing in size: larger samples is scaled down to a given number of individuals that permits the comparison of the numbers of species between samples differing in size: ( )E S N N n N n n i i S = − − = 1 1 , where , where E(Sn) = the expected number of species in a random sample of n individuals, S = the total number of species in the entire collection, Ni = the number of individuals per species i, N = the total number of individuals in the collection, and n = the sample size (number of individuals) chosen for standardization (see Heck et al. 1975, Krebs 1999). Sample size was scaled down to 6500 individuals (the number of individuals was 6839 in the north- west), and standard deviations were calculated for each sample size. For the analysis, data were pooled across fields. Results  Frequency of occurrence Certain species such as Capsella bursa-pastoris, Fumaria officinalis, Lamium spp., Plantago ma- jor, Stachys palustris and Thlaspi arvense were found only in the southern coastal region while others, i.e. Achillea millefolium, Achillea ptarmi- ca, Leontodon autumnalis and Rumex spp. oc- curred only in the northwestern coastal region (Ta- ble 2). The weed species Brassica spp., Galium spp., Lapsana communis, Myosotis arvensis, Poly- gonum aviculare, Tripleurospermum inodorum, Vicia spp. and Viola arvensis were more common in the south, while Galeopsis spp., Persicaria spp. and Spergula arvensis were more common in the northwest (Table 2). Among the perennial weed species, Elymus repens was somewhat more fre- quent in the northwest, while Cirsium arvense and Sonchus arvensis were more frequent in the south (Table 2). Chenopodium spp., E. repens and Stel- laria media were very common in both regions (Table 2). Number of weed species A total of 38 weed species and weed taxa were found in the coastal regions. Both the observed (SOBS) and expected (E(S6500)) total number of spe- cies were higher in the south than in the northwest: 33 and 31.8 (SD = 0.91) vs. 26 and 25.9 (SD = 0.23), respectively. The same pattern was found with regard to the average number of species per field. The average number of species in the south and northwest was 15.6 vs. 10.0 (median = 16.0 vs. 9.5; SD = 3.2 vs. 3.2), respectively (Z = –3.63; P < 0.001). Some of the species that were found are considered rare, i.e. Anchusa arvensis, Centaurea cyanus and Erodium cicutarium or extinct, i.e. Pa- paver dubium (Hämet-Ahti et al. 1998, Rassi et al. 2001). The number of weed species was not de- pendent on the duration of organic farming in ei- ther region (south: R2 = 0.11, P = 0.186; northwest: R2 = 0.005, P = 0.835). Abundance of single weed species The most frequent species were generally most abundant with respect to density and dry weight. The only exception was Sonchus arvensis, which was more prominent in the northwest with respect to dry weight than could be deduced from frequen- cies and densities. Weed abundance in the south differed from the northwest with respect to density and dry weight. The densities of Lapsana commu- nis, Myosotis arvensis, Polygonum aviculare, Son- chus arvensis, Tripleurospermum inodorum and Vicia spp. were significantly higher in the south, whereas E. repens, Persicaria spp. and Spergula arvensis (at the limit of significance) occurred at significantly higher densities in the northwest (Ta- ble 3). Dry weights of L. communis, Myosotis ar- vensis (at the limit of significance), Polygonum aviculare, Tripleurospermum inodorum and Vicia spp. (at the limit of significance) were significantly higher in the south, while E. repens, Galeopsis 172 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Riesinger, P. & Hyvönen, T. Weed occurrence in coastal regions of Finland Table 2. Frequencies of all weed species and weed taxa found in the southern and the northwestern coastal regions. Weed species and taxa BAYER code a Both regions Southern coastal region Northwestern coastal region Achillea millefolium L. ACHMI 3.3 – 8.3 Achillea ptarmica L. ACHPT 3.3 – 8.3 Anchusa arvensis (L.) MB ANCHA 3.3 5.6 – Brassica L. spp. BRASS 20.0 27.8 8.3 Capsella bursa-pastoris (L.) MEDIK. CAPBP 23.3 38.9 – Centaurea cyanus L. CENCY 3.3 5.6 – Chenopodium L. spp. CHESS 96.7 100.0 91.7 Cirsium arvense (L.) SCOP. CIRAR 16.7 22.2 8.3 Elymus repens (L.) GOULD AGRRE 83.3 77.8 91.7 Equisetum arvense L. EQUAR 26.7 33.3 16.7 Erodium cicutarium (L.) L HÈR. EROCI 3.3 5.6 – Erysimum cheiranthoides L. ERYCH 73.3 72.2 75.0 Fallopia convolvulus (L.) Á. LÖVE POLCO 60.0 66.7 50.0 Fumaria officinalis L. FUMOF 43.3 72.2 – Galeopsis L. spp. GAESS 76.7 66.7 91.7 Galium L. spp. GALSS 56.7 66.7 41.7 Gnaphalium uliginosum L. GNAUL 3.3 – 8.3 Grasses other than Elymus repens (L.) GOULD b – 46.7 27.8 75.0 Lamium L. spp. LAMSS 43.3 72.2 – Lapsana communis L. LAPCO 56.7 88.9 8.3 Leontodon autumnalis L. LEBAU 3.3 – 8.3 Myosotis arvensis (L.) HILL MYOAR 50.0 72.2 16.7 Papaver dubium L. PAPDU 3.3 5.6 – Persicaria L. spp. POLLA 63.3 44.4 91.7 Plantago major L. PLAMA 6.7 11.1 – Polygonum aviculare L. POLAV 56.7 72.2 33.3 Ranunculus repens L. RANRE 20.0 11.1 33.3 Rumex L. spp. RUMSS 6.7 – 16.7 Sonchus arvensis L. SONAR 56.7 77.8 25.0 Spergula arvensis L. SPRAR 66.7 61.1 75.0 Stachys palustris L. STAPA 3.3 5.6 – Stellaria media (L.) VILL. STEME 96.7 100.0 91.7 Taraxacum L. spp. TAROF 23.3 33.3 8.3 Thlaspi arvense L. THLAR 16.7 27.8 – Trifolium L. spp. b TRFSS 66.7 66.7 66.7 Tripleurospermum inodorum SCH. BIP. MATSS 53.3 72.2 25.0 Tussilago farfara L. TUSFA 3.3 5.6 – Veronica L. spp. VERSS 3.3 5.6 – Vicia L. spp. VICSS 30.0 44.4 8.3 Viola arvensis MURRAY VIOAR 76.7 88.9 58.3 a Bayer codes (BAYER 1992). b Not included in the number of weed species. 173 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 166–182. Table 3. Mean densities and dry weights of the weed species dominating in the southern and the northwestern coastal regions including statistical differences between regions. Density (shoots m-2) Z a P-value Dry weight (kg ha-1) Z a P-value South Northwest South Northwest Brassica spp. 5.6 0.1 –1.36 0.173 61.6 0.2 –1.39 0.164 Chenopodium spp. 156.3 99.8 –1.06 0.290 171.9 299.7 1.25 0.212 Cirsium arvense 0.6 0.2 –0.95 0.344 3.2 0.1 –1.01 0.312 Elymus repens 56.1 212.1 2.25 0.025 101.6 882.3 2.61 0.009 Erysimum cheiranthoides 23.5 17.4 1.43 0.152 13.1 20.6 1.62 0.104 Fallopia convolvulus 27.7 2.5 –1.51 0.131 45.1 7.0 –1.36 0.175 Fumaria officinalis 5.3 – –3.64 0.001 11.6 – –3.42 0.002 Galeopsis spp. 20.4 32.7 1.60 0.110 29.1 93.3 2.62 0.009 Galium spp. 10.0 4.4 –1.39 0.174 23.7 6.4 –1.67 0.108 Lamium spp. 24.1 – –3.63 <0.001 31.6 – –3.42 0.001 Lapsana communis 33.0 0.3 –4.02 <0.001 15.9 0 –4.13 <0.001 Myosotis arvensis 13.6 0.25 –3.12 0.004 3.9 0.4 –2.01 0.054 Persicaria spp. 6.2 19.8 2.07 0.038 5.4 32.9 2.76 0.006 Polygonum aviculare 4.5 0.9 –2.19 0.036 4.1 1.3 –2.05 0.049 Sonchus arvensis 19.7 3.7 –2.78 0.005 35.6 38.2 2.39 0.020 Spergula arvensis 19.7 120.0 1.96 0.050 5.7 122.1 2.02 0.044 Stellaria media 50.2 28.3 –1.74 0.082 43.3 24.1 –0.91 0.363 Tripleurospermum inodorum 16.6 0.3 –2.93 0.007 10.5 0.03 –2.74 0.011 Vicia spp. 6.8 0.3 –2.01 0.044 22.9 0.3 –2.04 0.051 Viola arvensis 43.5 16.9 –0.68 0.495 10.6 6.9 –0.30 0.765 a Wilcoxon test statistic. spp., Persicaria spp., Sonchus arvensis and Sper- gula arvensis had higher dry weights in the north- west (Table 3). More weed species dominated in the south than in the northwest. In the south, eight weed species and taxa (Chenopodium spp., E. repens, S. media, V. arvensis, L. communis, F. convolvulus, Lamium spp., E. cheiranthoides; ranked in descending or- der) made up 73.3% of total weed density, while in the northwest only three species and taxa (E. re- pens, Spergula arvensis, Chenopodium spp.; ranked in descending order) were required to reach 75.7% of total weed density (Fig. 1). 70.4% of the total weed dry weight in the south consisted of seven species and taxa (Chenopodium spp., E. repens, Brassica spp., F. convolvulus, S. media, Sonchus arvensis, Lamium spp.; ranked in descending order), while only two species (E. re- pens, Chenopodium spp.; ranked in descending order) made up 74.1% of the total weed dry weight in the northwest (Fig. 2). Abundance of crop and weeds Crop density was significantly higher in the south than in the northwest whereas crop dry weight did not differ between the regions (Table 4). Density of annual weed species did not differ between south and northwest whereas perennial weeds occurred at significantly higher density in the northwest (Table 4). Due to the high abundance of perennial species, weed dry weight was significantly higher in the northwest than in the south (Table 4). 174 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Riesinger, P. & Hyvönen, T. Weed occurrence in coastal regions of Finland 0 5 10 15 20 25 30 35 40 A G R R E C H E S S E R Y C H LA M S S LA P C O P O LC O S P R A R S T E M E V IO A R O th er s Proportion (% of weed density) South Northwest Fig. 1. Proportional density of the weed species and taxa dominating in the southern and the northwestern coastal regions. AGRRE = Elymus repens, CHESS = Chenopodi- um spp., ERYCH = Erysimum cheiranthoides, LAMSS = Lamium spp., LAPCO = Lapsana communis, POLCO = Fallopia convolvulus, SPRAR = Spergula arvensis, STEME = Stellaria media, VIOAR = Viola arvensis 0 10 20 30 40 50 60 A G R R E B R A S S C H E S S LA M S S P O LC O S O N A R S T E M E O th er s Proportion (% of weed dry weight) South Northwest Fig. 2. Proportional dry weight of the weed species and taxa dominating in the southern and the northwestern coastal regions. AGRRE = Elymus repens, BRASS = Brassica spp., CHESS = Chenopodium spp., LAMSS = Lamium spp., POLCO = Fallopia convolvulus, SONAR = Sonchus arvensis, STEME = Stellaria media Table 4. Mean densities and dry weights of crop and weeds in the southern and the northwestern coastal regions including statistical differences between regions. South Northwest Mean Median SD Mean Median SD Z a P-value Density (shoots m-2) Crop 685.1 657.5 205.5 527.7 505.5 153.2 –2.27 0.026 Weeds all species 565.4 467.5 437.4 569.9 507.5 334.4 0.32 0.751 annuals 483.1 363.5 399.5 344.2 255.5 264.8 –1.12 0.236 perennials 82.3 44.0 115.6 225.8 190.0 196.5 2.18 0.029 Dry weight (kg ha-1) Crop 6183.0 6564.7 2112.0 6685.4 6533.1 2390.2 0.15 0.882 Weeds all species 696.7 480.4 573.0 1594.1 1476.5 974.0 2.65 0.008 annuals 521.7 328.1 504.4 615.4 494.3 379.2 1.55 0.122 perennials 175.0 98.1 189.3 978.8 758.9 868.9 2.60 0.009 a Wilcoxon test statistic SD = standard deviation Average weed density in relation to weed and crop density was 41.6 vs. 48.7% in the south com- pared with the northwest while the average propor- tion of weed dry weight in relation to total weed and crop dry weight was 11.4 vs. 20.6%, respec- tively. The high abundance of perennial weed spe- cies in the northwest was due to E. repens. On av- erage, E. repens constituted 4.3 vs. 18% of the total crop and weed density and 1.8 vs. 10.8% of the average total crop and weed dry weight in the south 175 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 166–182. and northwest, respectively. In some crop stands E. repens was not found at all while in certain crop stands it made up close to 30% of the total dry weight. Distribution of weed abundance The spring cereal stands included in this survey differed remarkably from each other with regard to weed density and dry weight. In 30% of the inves- tigated spring cereal stands, weed densities ranged between 100–299 shoots m-2. In one third of the cereal fields, the density of weeds amounted to 300–599 shoots m-2 whereas it exceeded 600 shoots m-2 in 36.7% of the crop stands. With regard to lower classes of weed density, the distribution of spring cereal fields was about the same in both re- gions. However, the highest levels of weed density were found in the south (Fig. 3). The dry weight of weeds exceeded 500 kg ha-1 in 63.3% of the investigated crop stands. In the south, weed dry weight in half of the fields was lower than 500 kg ha-1 while it exceeded 1000 kg ha-1 in 27.8% of the fields. In the northwest, weed dry weights lower than 500 kg ha-1 were found in only 16.7% of the spring cereal stands. Weed dry weight exceeded 1000 kg ha-1 in 66.7% of the fields surveyed in the northwest (Fig. 4). In the south, weed dry weight tended to be de- pendent (R2 = 0.21, P = 0.055), and weed density was dependent (R2 = 0.23, P = 0.042) on the dura- tion of organic farming. In the northwest, no de- pendence was detected (weed dry weight: R2 = 0.02, P = 0.650; weed density: R2 = 0.029, P = 0.598). Discussion  Number of weed species in   the coastal regions The number of weed species found in the coastal regions was lower compared with the weed sur- veys conducted in the late 1990s (126 species found in 165 organically cropped fields; Salonen et al. 2001b), and in the early 1960s (304 species found in 2710 conventionally cropped fields; Mukula et al. 1969). The investigation by Mukula et al. (1969) covered all agricultural regions of 0 10 20 30 40 100-299 300-599 600-899 900- 1199 1200- 1499 1500- 1799 Weed density (shoots m-2) Proportion (% of stands) South Northwest Fig. 3. Distribution of cereal stands in the southern and the northwestern coastal regions with regard to weed density. 0 10 20 30 40 50 60 100- 499 500- 999 1000- 1499 1500- 1999 2000- 2499 2500- 2999 3000- 3499 Weed dry weight (kg ha-1) Proportion (% of stands) South Northwest Fig. 4. Distribution of cereal stands in the southern and the northwestern coastal regions with regard to weed dry weight. 176 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Riesinger, P. & Hyvönen, T. Weed occurrence in coastal regions of Finland Finland whereas the survey by Salonen et al. (2001b) included southern and central Finland. Since the number of species is dependent on the sample size (Heck et al. 1975), the likely reasons for the lower species number in the present study were the lower number of studied fields and the restriction to coastal regions. Furthermore, in our investigation, all weed species were not distin- guished by species level, but instead pooled by genera and taxa, which underestimates the real number of weed species. Despite of the small sample size, some rare weed species were encountered. This is in accord- ance with previous findings by Albrecht and Mat- theis (1998). Thus, organic cropping benefits bio- diversity by the conservation of rare and endan- gered species. The number of weed species was higher in the south than in the northwest, which is in agreement with previous investigations (Mukula et al. 1969, Salonen et al. 2001b). Weed species number in the south is enhanced by a longer grow- ing season and a larger species pool (Hämet-Ahti et al. 1998). In agreement with Becker and Hurle (1998), the number of weed species was not relat- ed to the duration of organic farming. Rather than the duration of organic farming, the factors that de- termine the number of weed species include single crop management measures, such as choice of crop rotation, soil tillage, weed management, and ferti- lization. In comparison with the surveys carried out by Mukula et al. (1969) and Salonen et al. (2001b), the number of dominating weed species was on the same level in the south, while it was lower in the northwest. About 70% of the total weed density had been made up by eight species in convention- ally cropped spring cereals in the beginning of the 1960s (Mukula et al. 1969) and by ten species in organically cropped stands 1984–1986 (Mela 1988). This proportion was reached by the accu- mulated density of eight weed species in the south whereas in the northwest only three species to- gether exceeded 70% of total weed density. Eight species had constituted 70% of the total weed dry weight recorded 1961–1964 (Mukula et al. 1969). In our investigation, the seven most dominant weed species in the south reached 70% of the total weed dry weight. In the northwest, however, this propor- tion of weed dry weight was made up of only two species. In the south, milder winters offer better conditions for cropping of winter cereals and thus for more diverse crop rotations whereas in the northwest fields are subjected to more simple sum- mer cereal or ley-and-cereal rotations. Therefore, a lower number of weed species gain dominant posi- tions in the northwest than in the south (Mukula et al. 1969, Hald 1999). Weed communities in the coastal regions In agreement with previous investigations (Mukula et al. 1969, Salonen et al. 2001b), regional differ- ences in the occurrence of weed species were de- tected. Species preferring a warm climate such as Lamium spp. and S. palustris were only found in the south (Mukula et al. 1969, Salonen 1993). Lamium spp. and F. officinalis are known to thrive in dry clay soils (Mukula et al. 1969, Salonen 1993, Hallgren 1996). Clay soils prevail in the south, which might contribute to the fact that Lam- ium spp. and F. officinalis were found only in this region. Likewise, the significantly higher densities and dry weights of M. arvensis, P. aviculare and T. inodorum in the south can be explained by the preference of these weed species for dry mineral soils (Mukula et al. 1969, Erviö and Salonen 1987, Ellenberg et al. 1991, Erviö et al. 1994). On the other hand, the significantly higher abundances of Galeopsis spp. (by dry weight), Persicaria spp. and Spergula arvensis (by both density and dry weight) in the northwest are likely due to the moist coarse mineral and organic soils common in this region (Mukula et al. 1969, Salonen 1993). The tolerance of Persicaria spp. and Spergula arvensis to soils with low pH results in significantly higher densities and dry weights of these species in the northwest where such soils are more common than in the south. In the contrary, significantly higher abundances of weed species preferring soils with high pH, such as Lamium spp., L. communis and M. arvensis were found on the clay soils of the south (Albrecht and Bachthaler 1989, Erviö et al. 1994). 177 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 166–182. Galeopsis spp., Persicaria spp. and Spergula arvensis are classified as summer annuals, while A. arvensis, C. cyanus, E. cicutarium, M. arvensis and P. dubium also readily germinate in autumn (Fogelfors 1977). The latter are favoured by the cultivation of winter annual crops, which is com- mon in the south, but rather an exception in the northwest (Table 1). Brassica spp. was more fre- quent in the south than in the northwest, which was not expected since Salonen et al. (2001b) have not found any consistent differences between locali- ties with regard to the occurrence of this weed taxon. Brassica spp. disseminates independently of geographical region as a volunteer crop, or through seeds contained in feed cereals that trans- fer to manure fertilizer. A probable reason for the relatively low occurrence of Brassica spp. in the northwest is the lower proportion of turnip rape both in conventional and organic crop rotations ap- plied in this region (Information Centre of the Ministry of Agriculture and Forestry 2003). In contrast to the results obtained by this sur- vey, C. bursa-pastoris, F. officinalis and T. arvense have previously been recorded as frequent in the northwest (Mukula et al. 1969, Mela 1988, Salo- nen et al. 2001b). G. uliginosum, which was found in the northwest only, has been reported to occur in the south, too, although at lower frequencies than in the northwest (Mukula et al. 1969, Mela 1988, Salonen et al. 2001b). The cause for these contra- dictions is probably the low number of fields in- cluded in this survey. Most of the annual weed species and seedlings of perennial species that were found in the spring cereal stands in this survey can be controlled by weed harrowing. Exceptions are Galeopsis spp. (high dry weight in the northwest) and Galium spp. (frequent both in the northwest and the south). Weed harrowing is effective against the seedling and early rosette stages of weeds and can be car- ried out before the emergence of the crop, as well as after DC 13 (Habel 1954, Koch 1959, Kees 1962). In contrast to weed harrowing, stubble cul- tivation is only effective against those weed spe- cies germinating in autumn (Boström and Fogel- fors 1999). Stubble cultivation is effective against volunteer crops as well, although in the case of B. rapa spp. oleifera, cultivation must be shallow (Pekrun and Lutman 1998). The perennial weed species E. repens, C. ar- vense and Sonchus arvensis have been regarded as the most noxious for organic cropping in Finland (Salonen et al. 2001b). C. arvense and Sonchus ar- vensis occurred at higher frequencies in the south compared with the northwest. The density of Son- chus arvensis was significantly higher in the south, while its dry weight was significantly higher in the northwest. Higher occurrences of these weeds in the south have previously been recorded by Muku- la et al. (1969), Mela (1988) and Salonen et al. (2001b). C. arvense prefers clay soils while Son- chus arvensis also thrives in coarse mineral soils (Mukula et al. 1969, Salonen 1993). C. arvense has a deep reaching root system and therefore competes successfully with agricultural crops, es- pecially in the dry clay soils that prevail in the south. The establishment of both C. arvense and Sonchus arvensis is favoured by a high proportion of annual crops (Mukula et al. 1969, Donald 1990) while cultivation of ley in combination with three cuts is an appropriate measure against C. arvense (Dock-Gustavsson 1997). On farms where the pro- portion of ley is low, C. arvense and Sonchus ar- vensis can be controlled by stubble cultivation, provided that tillage operations can be launched in early autumn (Boström and Fogelfors 1999). The higher abundance of E. repens in the north- west is in line with the results obtained by the lat- est survey of inland sites (Salonen et al. 2001b) and is explained by the preference of this weed for moist silt and organic soils (Mukula et al. 1969, Erviö et al. 1994). Compared with the 1960s, the frequency of E. repens in fields not treated with herbicides has generally increased (cf. Mukula et al. 1969, Mela 1988, Salonen 2001b). E. repens has an extensive rhizome system and its spreading is incited by lower tillage intensity (Håkansson 2003). Stubble cultivation accomplished twice per autumn is recommended as an efficient measure against E. repens (Boström and Fogelfors 1999). However, the further to the north the fields are situ- ated, the more limited are the opportunities for stubble cultivation after the harvest of grain crops. Cropping of late-ripening varieties further short- 178 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Riesinger, P. & Hyvönen, T. Weed occurrence in coastal regions of Finland ens the period suitable for stubble cultivation. E. repens had previously been less frequent and less abundant in the northwest than in the south (Muku- la et al. 1969, Mela 1988). More rare and less fa- vourable opportunities for stubble cultivation than in the south have likely favoured the spreading of E. repens in the northwest. Due to the absence of tillage, the maintenance of leys for more than three years favours the spreading of E. repens (Håkans- son 2003). Differences in the occurrence of weed species between the southern and the northwestern coastal regions can thus be related to climate and edaphic as well as to management factors. Owing to the occurrence of different weed species, appropriate weed management in the south and north must rely on different strategies. In regions with short grow- ing seasons or on organic soils, perennial weed species emerging from root shoots and rhizomes are generally restrained by integrating a spring and early summer fallow into the regular crop rotation (Bylterud 1965, Kakriainen-Rouhiainen et al. 2003). Weed abundance and  cropping intensity Compared with previous national surveys, Cheno- podium spp. in organically cropped stands reached higher abundances both in the southern and the northwestern coastal regions. The abundances of Brassica spp., F. convolvulus, Galium spp. and Lamium spp. in the south were much higher com- pared with previously found national average num- bers. In the northwest, E. repens exceeded previ- ously recorded average levels of density and dry weight (Mukula 1974, Mela 1988, Salonen et al. 2001b). The density of E. cheiranthoides was the same as reported at national level in the 1960s (Mukula 1974) and 1980s (Mela 1988). However, its dry weight in the northwest was twice as high. The above mentioned weed species are either nitrophilous or, owing to their climbing growth form, competitive in conditions of increasing ni- trogen levels (Mahn 1988, Albrecht and Bachthaler 1989, Ellenberg et al. 1991, Jørnsgård et al. 1996, Blackshaw et al. 2003). Of the weed species found to be more abundant in the coastal regions com- pared with previously recorded national average levels only C. bursa-pastoris, Spergula arvensis and Vicia spp. are non-nitrophilous (Ellenberg et al. 1991). These findings suggest higher nitrogen levels compared with the organically cropped fields investigated by previous surveys (Mela 1988, Salonen et al. 2001b) as well as compared with the conventionally cropped fields included in the first national weed survey 1961-1964 (Mukula et al. 1969). The dominance of nitrophilous weed spe- cies in organically cropped spring cereal stands in the coastal regions of Finland contradicts the pre- diction by Rydberg and Milberg (2000) that con- version to organic farming practices would lead to an increase of non-nitrophilous weed species. High nitrogen levels in organically cropped fields might be due to high humus levels, to frequent cropping of legumes, to recent application of manure to a single field or to regular import of nitrogen by ma- nure acquired from other farms. High soil nitrogen levels offer a potential for high crop yields, pro- vided that weeds are kept under control. Frequencies of ruderal and perennial grassland weeds such as A. millefolium, A. ptarmica, E. ar- vense, E. cicutarium, L. autumnalis, P. major, R. repens, Rumex spp. and T. farfara were somewhat lower compared with the average national frequen- cies reported by Salonen et al. (2001b) and Mela (1988), and much lower than the frequencies found in the beginning of the 1960s (Mukula 1974). The decrease of these species has been attributed to land amelioration, the decline of agricultural land devoted to pasture, higher shares of annual crops, shorter duration of leys and intensified soil tillage, as well as to the application of herbicides (Erviö and Salonen 1987, Rydberg and Milberg 2000). Owing to the fundamental importance of symbi- otic nitrogen fixation by red clover-grass leys for the nitrogen supply of non-leguminous crops, grassland is part of the regular crop rotation, espe- cially on organic farms. The leys are usually termi- nated after two or three harvest years, tilled and grown with annual crops. A spreading of ruderal and perennial grassland weeds under organic farm- 179 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 166–182. ing management is therefore not to be expected. Important exceptions are C. arvense and E. repens which are adapted to soil cultivation. Consequent- ly, the frequencies and abundances of these species have increased. Weed density in the coastal regions was on about the same level as reported in previous na- tional surveys of low-input conventional and or- ganic farming (Mukula et al. 1969: 550 shoots m-2; Mela 1988: 505 shoots m-2; Salonen et al. 2001b: 469 shoots m-2). Average weed dry weight found in the south (697 kg ha-1) exceeded the national aver- age found in the 1980s (Mela 1988: 575 kg ha-1), but was on the same level as found by the recent national survey of organically cropped spring ce- real fields (Salonen et al. 2001b: 678 kg ha-1). In the northwest, however, weed dry weight was tre- mendously high, reaching an average of 1594 kg ha-1, and thus widely exceeding the national aver- age of 1000 kg ha-1 found in the 1960s (Mukula et al. 1969). The difference in weed dry weight be- tween south and northwest is supported by Salo- nen et al. (2001b) who had found higher weed dry weight in the western part of central Finland com- pared with the southern mainland. The significant- ly higher crop density in the southern coastal re- gion probably contributed to higher competitive- ness of the crop and thus to lower weed abundance (Erviö 1972, Erviö 1983). While the level of weed infestation was clearly influenced by climate, edaphic factors and management, there was only a weak positive correlation between the duration of organic farming and weed abundance in the south, and no correlation at all in the northwest. Regard- less of region, the investigated spring cereal crops differed widely from each other with respect to weed abundance. Conclusions  Weed communities of organically cropped spring cereal stands in the southern and the northwestern coastal regions differed from each other with re- gard to species composition, frequency of occur- rence and abundance. Weed species frequencies and abundances in the coastal regions differed from the inland regions. By favouring populations of rare and threatened weed species, organic farm- ing proved to be beneficial for species diversity. The present level of weed occurrence in organ- ically cropped stands of spring cereals in the coast- al regions calls for target-oriented weed manage- ment over an extended time span as well as the integration of preventive, cultural and physical control measures. Due to different weed communi- ties, different weed control measures have to be applied in the southern, and in the northwestern coastal region, respectively. Acknowledgements. We thank the R. Erik Serlachius foun- dation for supporting the sampling, the Göran and Louise Ehrnrooth foundation for supporting the evaluation of data and Jukka Salonen for comments on the manuscript. References Albrecht, H. & Bachthaler, G. 1989. Unkrautsamengehalte und Bodeneigenschaften von Ackerflächen in Bayern. Weed Research 30: 101–108. Albrecht, H. & Mattheis, A. 1998. The effects of organic and integrated farming on rare arable weeds on the For­ schungsverbund Agrarökosysteme München (FAM) research station in southern Bavaria. Biological Con- servation 86: 347–356. Bayer, 1992. Important crops of the world and their weeds. 2nd ed. Leverkusen, Germany: BAYER AG. 1682 p. Becker, B. & Hurle, K. 1998. Unkrautflora auf Feldern mit unterschiedlich langer ökologischer Bewirtschaftung. Zeitschrift für Pflanzenkrankheiten und Pflanzenschutz. Sonderheft 16: 155–161. (in German) Blackshaw, R.E., Brandt, R.N., Janzen, H.H., Entz, T., Grant, C.A. & Derksen, D. 2003. Differential response of weed species to added nitrogen. Weed Science 51: 532–539. Boström, U. & Fogelfors, H. 1999. Type and time of autumn tillage with and without herbicides at reduced rates in Southern Sweden. Soil & Tillage Research 50: 283– 293. Bylterud, A. 1965. Mechanical and chemical control of Agro- pyron repens in Norway. Weed Research 5: 169–180. Cardina, J., Regnier, E. & Harrison, K. 1991. Long­term till­ age effects on seed banks in three Ohio soils. Weed Science 39: 186–194. Davies, D.H.K., Christal, A., Talbot, M., Lawson, H.M. & Wright, G.McN. 1997. Changes in weed populations in the conversion of two arable farms to organic farming. 180 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Riesinger, P. & Hyvönen, T. Weed occurrence in coastal regions of Finland In: Proceedings of the 1997 Brighton Crop Protection Conference – Weeds. Brighton, UK. p. 973–978. Dock­Gustavsson, A.­M. 1997. Growth and regenerative capacity of plants of Cirsium arvense. Weed Research 37: 229–236. Donald, W.W. 1990. Management and control of Canada thistle (Cirsium arvense). Reviews of Weed Science 5: 193–250. Doucet, C., Weaver, S.E., Hamill, A.S. & Zhang, Z. 1999. Separating the effects of crop rotation from weed man­ agement on weed density and diversity. Weed Science 47: 729–735. Ellenberg, H., Weber, H.E., Düll, R., Wirth, W., Werner, W. & Paulißen, D. 1991. Zeigerwerte von Pflanzen in Mittel­ europa. Scripta Geobotanica 18: 1–248. El Titi, A. 1995. Ecological aspects of integrated farming. In: Glen, D.M. et al. (eds.). Ecology and integrated farming systems. New York, USA: John Wiley & Sons. p. 243– 256. Erviö, L.­R. 1972. Growth of weeds in cereal populations. Journal of the Scientific Agricultural Society in Finland 44: 19–28. Erviö, L.­R. 1983. Competition between barley and annual weeds at different sowing densities. Annales Agricultu- rae Fenniae 22: 232–239. Erviö L.­R. & Salonen, J. 1987. Changes in the weed popu­ lation of spring cereals in Finland. Annales Agriculturae Fenniae 26: 201–226. Erviö, R., Hyvärinen, S., Erviö, L.­R. & Salonen, J. 1994. Soil properties affecting weed distribution in spring ce­ real and vegetable fields. Agricultural Science in Fin- land 3: 497–504. European Council 1991. European Council Regulation (EC) No. 2092/91. Available on the Internet: http://europa. eu.int/eur­lex/en/consleg/ main/1991/en_1991R2092_ index.html European Council 1999. European Council Regulation (EC) No. 1804/1999. Available on the Internet: http://europa. e u . i n t / e u r ­ l e x / p r i / e n / o j / d a t / 1 9 9 9 / l _ 2 2 2 / l _ 22219990824en00010028.pdf Finnish Meteorological Institute 2003. Climate data. Helsin­ ki, Finland. Fogelfors, H. 1977. Åkerogräs i Sverige. Lantbrukshögsko- lans meddelanden B 19. Uppsala, Sweden. 233 p. (in Swedish) Freyer, B. 1991. Ökologischer Landbau – Planung und Ana- lyse von Betriebsumstellungen. Ökologie & Landwirt­ schaft 2. Weikersheim, Germany: Verlag Josef Mark­ graf. 229 p. (in German) Frick, B. & Thomas, A.G. 1992. Weed surveys in different tillage systems in southwestern Ontario field crops. Ca- nadian Journal of Plant Science 72: 1337–1347. Habel, W. 1954. Über die Wirkungsweise der Eggen gegen Samenunkräuter sowie die Empfindlichkeit der Un- krautarten und ihrer Altersstadien gegen den Eggvor- gang. Dissertation. Hohenheim, Germany. 57 p. (in German) Håkansson, S. 2003. Weeds and weed management on ar- able land. Wallingford, UK: CABI Publishing. 274 p. Hald, A.B. 1999. Weed vegetation (wild flora) of long estab­ lished organic versus conventional cereal fields in Den­ mark. Annals of Applied Biology 134: 307–314. Hallgren, E. 1996. Occurrence of broad­leaved weeds on different soils in different crops in Sweden. Swedish Journal of Agricultural Research 26: 115–123. Hämet­Ahti, L., Suominen, J., Ulvinen, T. & Uotila, P. 1998. Retkeilykasvio (Field Flora of Finland). Helsinki, Fin­ land: Finnish Museum of Natural History, Botanical Mu­ seum. 656 p. (in Finnish) Heck, K.L.J., van Belle, G. & Simberloff, D. 1975. Explicit calculation of the rarefaction diversity measurement and the determination of sufficient sample size. Ecolo- gy 56: 1459–1461. Hyvönen, T. & Salonen, J. 2001. Weed species diversity and community composition in cropping practices at two intensity levels – a six­year experiment. Plant Ecology 159: 73–81. Hyvönen, T., Ketoja, E., Salonen, J., Jalli, H. & Tiainen, J. 2003. Weed species diversity and community composi­Weed species diversity and community composi­ tion in organic and conventional cropping of spring ce­ reals. Agriculture, Ecosystems and Environment 97: 131–149. Information Centre of the Ministry of Agriculture and For­ estry (Maa­ ja metsätalousministeriön tietopalvelukes­(Maa­ ja metsätalousministeriön tietopalvelukes­ kus) 2003. Maatilarekisteri 2002 (Farm Register 2002). Helsinki, Finland. 130 p. Jensen, E.S. 1986. Symbiotic N2 fixation in pea and field bean estimated by 15N fertilizer dilution in field experi­ ments with barley as a reference crop. Plant and Soil 92: 3–13. Jørnsgård, B., Rasmussen, K., Hill, J. & Christiansen, J.L. 1996. Influence of nitrogen on competition between ce­ reals and their natural weed populations. Weed Rese- arch 36: 461–470. Kakriainen­Rouhiainen, S., Väisänen, J., Vanhala, P. & Löt­ jönen, T. 2003. Mid­summer bare fallow effective inMid­summer bare fallow effective in controlling perennial weeds. In: Niemeläinen, O. & Topi­Hulmi, M. (eds.). Nordic agriculture in global per- spective. Proceedings of the NJF’s 22nd Congress. Tampere, Finland: Tampere University Print. p. 104. Kees, H. 1962. Untersuchungen zur Unkrautbekämpfung durch Netzegge und Stoppelbearbeitungsmassnah- men unter besonderer Berücksichtigung des leichten Bodens. Dissertation. Hohenheim, Germany. 102 p. (in German) Koch, W. 1959. Untersuchungen zur Unkrautbekämpfung durch Saatpflege und Stoppelbearbeitungsmassnah- men. Dissertation. Hohenheim, Germany. 123 p. (in German) Krebs, C.J. 1999. Ecological methodology, Second edition. California, USA: Benjamin/Cummins. 620 p. Kurki, M. 1982. Suomen peltojen viljavuudesta III. Helsinki, Finland. 181 p. (in Finnish, with English summary) Leeson, J.Y., Sheard, J.W. & Thomas, A.G. 2000. Weed communities associated with arable Saskatchewan farm management systems. Canadian Journal of Plant Science 80: 177–185. Mäder, P., Fliessbach, A., Dubois, D., Gunst, L., Fried, P. & Niggli, U. 2002. Soil fertility and biodiversity in organic farming. Science 296: 1694–1697. Mahn, E.­G. 1988. Changes in the structure of weed com­ munities affected by agro­chemicals – what role does nitrogen play? Ecological Bulletins 39: 71–73. Mela, T. 1988. Luonnonmukainen peltoviljely Suomessa. 181 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 166–182. Viljelymenetelmät, rikkakasvit, peltojen viljavuus, sadot ja sadon laatu. English summary: Organic farming inEnglish summary: Organic farming in Finland. Cultivation methods, weeds, soil fertility, yields and yield quality. University of Helsinki. Department of Plant Husbandry. Publications 16. Helsinki, Finland. 220 p. Mukula, J., Raatikainen, M., Lallukka, R. & Raatikainen, T. 1969. Composition of weed flora in spring cereals in Finland. Annales Agriculturae Fenniae 8: 59–110. Mukula, J. 1974. Weed competition in spring cereal fields in Finland. Forskning og forsøk i lantbruket 25: 585–592. Mukula, J. & Rantanen, O. 1987. Climatic risks to the yield and quality of field crops in Finland. I. Basic facts about Finnish field crop production. Annales Agriculturae Fenniae 26: 1–18. National Board of Agriculture (Maatilahallitus) 1991.(Maatilahallitus) 1991. Maati- latilastollinen vuosikirja 1990 (Yearbook of Farm Statis­ tics 1990). Helsinki, Finland. 254 p. Nykänen, A., Granstedt, A., Laine, A. & Kunttu, S. 2000. Yields and clover contents of leys of different age in organic farming in Finland. Biological Agriculture and Horticulture 18: 55–66. Pekrun, C. & Lutman, P.J.W. 1998. The influence of post­ harvest cultivation on the persistence of volunteer oilseed rape. Aspects of Applied Biology 51: 113–118. Plant Production Inspection Centre (Kasvintuotannon tar­Kasvintuotannon tar­ kastuskeskus) 2003. Luonnonmukainen maatalous 2002 – Tilastoja (Organic Farming 2002 – Statistics). Julkaisuja B2 Luomutuotantu 2/2003. Loimaa, Finland. 56 p. Pyšek, P. & Lepš, J. 1991. Response of a weed community to nitrogen fertilisation: a multivariate analysis. Journal of Vegetation Science 2: 237–244. Rantzau, R., Freyer, B. & Vogtmann, H. 1990. Umstellung auf ökologischen Landbau. Schriftenreihe des Bundes­ ministers für Ernährung, Landwirtschaft und Forsten. Reihe A: Angewandte Wissenschaft 389. Münster­Hil­ trup, Germany: Landwirtschaftsverlag GmbH. 153 p. (in German) Rasmussen, K., Holst, N. & Kristensen, I.S. 1998. Weeds atWeeds at eight organic livestock farms – significant factors for weed dynamics 1989–1996. In: Danmarks Jordbrugs­ forskning. 15th Danish plant protection conference. Side effect of pesticides, weeds. Markbrug 2: 203–217. Foulum, Denmark. (in Danish) Rassi, P., Alanen, A., Kanerva, T. & Mannerkoski, I. (eds.). 2001. The 2001 red list of Finnish species. Ministry of the Environment and Finnish Environment Institute. Helsinki, Finland. Available on the Internet: http://www. vyh.fi/eng/environm/naturcon/threat/2000/2000.htm Rydberg, N.T. & Milberg, P. 2000. A survey of weeds in or­ ganic farming in Sweden. Biological Agriculture and Horticulture 18: 175–185. Salonen, J. 1993. Weed infestation and factors affecting weed incidence in spring cereals in Finland – a multi­ variate approach. Agricultural Science Finland 2: 525– 536. Salonen, J., Hyvönen, T. & Jalli, H. 2001a. Weeds in spring cereal fields in Finland – a third survey. Agricultural and Food Science in Finland 10: 347–364. Salonen, J., Hyvönen, T. & Jalli, H. 2001b. Weed flora in organically grown spring cereals in Finland. Agricultural and Food Science in Finland 10: 231–242. Sokal, R.R. & Rohlf, F.J. 1995. Biometry. The principles and practice of statistics in biological research. 3rd edition. New York, USA: W.H. Freeman and Company. 887 p. Svensson, R. & Wigren, M. 1986. Observations on the de­ cline of some farmland weeds. Memoranda Societas Pro Fauna Flora Fennica 62: 63–67. van Elsen, T. 2000. Species diversity as a task for organic agriculture in Europe. Agriculture, Ecosystems and En- vironment 77: 101–109. Vuorinen, J. & Mäkitie, O. 1955. The method of soil testing in use in Finland. Agrogeological Publications 63: 1– 44. Zadoks, J.C., Chang, T.T. & Konzak, G.F. 1974. A decimal code for the growth stage of cereals. Weed Research 14: 415–421. 182 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Riesinger, P. & Hyvönen, T. Weed occurrence in coastal regions of Finland SELOSTUS Rikkakasvien esiintyminen rannikkoalueiden luomukevätviljapelloilla Paul Riesinger ja Terho Hyvönen Helsingin yliopisto ja Maa- ja elintarviketalouden tutkimuskeskus Tutkimuksessa vertailtiin eteläisen (Itä-Uusimaa, Uusi- maa, Varsinais-Suomi ja Ahvenanmaa) ja Pohjanmaan rannikkoalueiden luonnonmukaisesti viljeltyjen kevät- viljapeltojen rikkakasviyhteisöjä. Alueet eroavat toisis- taan maataloustuotannon erikoistumisen, ilmaston ja maaperän suhteen. Näiden tekijöiden odotettiin aiheut- tavan eroja rikkakasviyhteisöjen lajimääriin sekä lajien esiintymisfrekvensseihin, tiheyksiin ja kuivapainoihin. Sekä lajien kokonaismäärä että keskimääräinen laji- määrä olivat suurempia eteläisellä rannikkoalueella kuin Pohjanmaalla (33 vs. 26 ja 17,3 vs. 10,8). Eteläiseltä rannikkoalueelta löydettiin joitakin harvinaisia lajeja (esim. ruisunikko), ja joitakin yleisiä lajeja tai sukuja, kuten peltoemäkki ja peipit, tavattiin ainoastaan etelästä. Linnunkaalin, peltolemmikin, pihatattaren, peltosaunion ja virnojen tiheydet ja kuivapainot olivat suuremmat ete- läisellä rannikkoalueella, kun taas juolavehnän, ukontat- tarien ja peltohatikan tiheydet ja kuivapainot olivat suu- remmat Pohjanmaan rannikolla. Rikkakasvien koko- naistiheys ei eronnut tutkimusalueiden välillä (565 vs. 570 versoa m-2). Sen sijaan kokonaiskuivapaino oli suu- rempi Pohjanmaalla kuin eteläisellä rannikkoalueella (1594 vs. 697 kg ha-1), mikä paljolti johtui suuresta juo- lavehnän määrästä Pohjanmaalla. Ainoastaan eteläisellä rannikkoalueella rikkakasvi- tiheys vaihteli luomuviljelyn keston mukaan. Aiempiin kartoituksiin verrattuna typpeä suosivat rikkakasvilajit olivat runsaampia, kun taas nurmipitoisessa viljelykier- rossa yleisten monivuotisten rikkakasvilajien runsaus oli alhaisempi. Tämä johtunee luomuviljelyn tehokkuuden noususta Suomessa. Rikkakasviyhteisöjen alueelliset ominaispiirteet pitäisi ottaa huomioon suunniteltaessa rikkakasvien torjuntaa. A G R I C U L T U R A L   A N D   F O O D   S C I E N C E A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Weed occurrence in Finnish coastal regions:a survey of organically cropped spring cereals Introduction Material and methods Results Discussion Conclusions References SELOSTUS