AGRICULTURAL AND FOOD SCIENCE

Agricultural and Food Science (2023) 32: 51–68

https://doi.org/10.23986/afsci.130009

Fifth survey on weed flora in spring cereals in Finland

Jukka Salonen, Heikki Jalli, Anne Muotila, Mari Niemi, Hannu Ojanen, Pentti Ruuttunen and Terho Hyvönen
Natural Resources Institute Finland (Luke), Tietotie 4, FI-31600 Jokioinen
email: [email protected]

Regular weed surveys provide information on changes in weed communities. The composition of weed flora in
spring cereal fields was investigated in southern and central Finland during 2020–2022. The survey was
conducted in 16 regions, 226 farms and 549 fields. The occurrence of weeds was assessed in the second half of
July from 90 organically cropped and 459 conventionally cropped fields. The weed flora was dominated by broad-
leaved species, representing 81% of the total number of 138 identified species. The average species number per
field was 21 in organic fields and 12 in conventional fields. The most frequent weed species in organic fields were
Chenopodium album 99%, Erysimum cheiranthoides 84%, Viola arvensis 83% and Stellaria media 82%. In
conventionally cropped fields, the most frequent species were Viola arvensis 83%, Stellaria media 65%, Galeopsis
spp. 59% and Galium spurium 59%. The two most common grass species in both production systems were Poa
annua and Elymus repens. The frequency and density of Poa annua had increased substantially in conventional
cropping since the previous survey in 2007–2009. The average density of weeds was 384 plants m -2 in organic
fields and 147 plants m-2 in sprayed conventional fields. The average air-dry biomass of weeds was 678 kg ha -1 and
151 kg ha-1, respectively. Elymus repens produced almost a quarter of the total weed biomass in both production
systems. Changes in weed flora were minor in terms of frequency and density of the most common broad-leaved
weed species. Increased infestation of Poa annua in conventional cropping calls for special attention. The
dominance of Elymus repens decreased in both production systems since the previous survey.

Key words: biodiversity, species diversity, conventional farming, organic farming, Elymus repens, Poa annua

Introduction
Arable weed communities are of agronomic, economic and ecological interest. Changes in agricultural practices
and environmental factors are reflected in weed abundance and species composition (Fried et al. 2008,
Andreasen and Streibig 2011, Pinke et al. 2012, Salonen et al. 2013). Therefore, periodic updating of the
information on weed communities is needed. This can be obtained by conducting regular weed surveys
(Krähmer et al. 2020).

National weed surveys have been conducted in many countries, including among others Spain (Cirujeda et al.
2011), Hungary (Novák et al. 2012), Germany (De Mol et al. 2015), Denmark (Andreasen et al. 2018) and
Slovakia (Májeková et al. 2019). The major aim of those surveys has usually been to get updated information on
the species composition of weed communities and the affecting factors. However, the focus can also be on the
effect of specific agricultural factors, such as conservation agriculture (Derrouch et al. 2022), crop diversification
(Hofmeijer et al. 2021) or incidence of herbicide resistance (Loureiro et al. 2017, Kolářová et al. 2023).
Monitoring weeds and delivering information on weed biology and their incidence in arable fields is helpful for
developing weed management strategies (Van Acker 2009, Borger et al. 2012).

In Finland, four extensive surveys of weeds in spring cereal fields have been carried out since the 1960s, the first
in 1961–1964 (Mukula et al. 1969), and thereafter in 1982–1984 (Erviö and Salonen 1987), in 1997–1999
(Salonen et al. 2001a) and in 2007–2009 (Salonen et al. 2011). Responding to an increasing interest in organic
production, organically cropped fields were included in the survey protocol from the 1990s survey onwards
(Salonen et al. 2001b). Weed surveys in spring cereal fields in Finland are important in terms of crop dominance
because spring cereals account for more than half of the total cropped area of around 2 million hectares (Luke
2023). The updated information on weed occurrence and diversity will be utilized by farmers, advisory services,
authorities and industry as well for scientific communication.

In each decade, a special focus has been included in the weed surveys in Finland. During the latest two surveys,
the development of production systems and cropping methods have been twofold regarding weed flora.
Increase in the area of organic cropping has extensified production, leading to promotion of weed flora, both in
terms of diversity and abundance (Rydberg and Milberg 2000, Hyvönen et al. 2003, Romero et al. 2008,
51
 J. Salonen et al.

Wortman et al. 2010). The area of organic production (incl. two-year transient period) in Finland has steadily
increased from 170 876 ha in 2010 to 339 460 ha in 2022, corresponding to the increase in the proportion of
cultivated

Received 16 May 2023 / Accepted 12 July 2023

The Scientific Agricultural Society of Finland
©This is an open access article under the CC BY 4.012
area from 7.5% to 15.0% (Ruokavirasto 2022). On the other hand, reduced tillage and direct drilling have partly
replaced ploughing in the 2000s, followed by an increase in the use of glyphosate, leading to reduction of
Elymus repens (Salonen et al. 2013). Common use of herbicides has brought new challenges such as resistance
in Stellaria media against sulfonylureas, as reported in Denmark already in the early 1990s (Kudsk et al. 1995)
and in Finland in 2013 (www.weedscience.com).

In this study, we report the results of the fifth weed survey of Finnish spring cereals, conducted during 2020–
2022. The present composition of weed flora is reported by focusing both on biodiversity and agronomic
aspects of weed occurrence.

We expected the above-described trends in cropping practices to affect the species composition of the weed
flora and the level of weed occurrence. Therefore, a divergent effect was expected regarding organically and
conventionally cropped fields. In both production systems, long-term impacts are of interest because the
current trends have been evident for decades.

Material and methods

Study regions, farms and fields
The weed survey was carried out in southern and central Finland in 2020–2022 (Fig. 1). The highest proportion
of fields (N = 306, 56%) was studied in southern Finland (Table 1), where spring cereals are grown on more than
50% of the arable area and annual spring-sown crops predominate (Luke 2023). The number of study fields
decreased towards the cattle farming regions in the east and north, where spring cereals account for less than
50% of the cultivated area and grassland is often included in the crop rotation. The number of farms visited was
226, of which 189 were engaged in conventional and 37 in organic production. We aimed at including 10–15%
organically grown fields in all survey regions. This corresponds to 16% of all studied fields and is close to the
proportion of organically cropped field area (15%) in Finland in 2022 (Ruokavirasto 2022).

Fig 1. Weed survey regions (1–16) in three zones in 2020–2022. Key to region numbers:

52
 Agricultural and Food Science (2023) 32: 51–68

1 = Jokioinen, 2 = Lammi, 3 = Nauvo/Korppoo, 4 = Tammela, 5 = Laukaa/Toivakka, 6 = Kitee, 7 =

Mikkeli, 8 = Lieto/Paimio, 9 = Laihia, 10 = Nivala, 11 = Laitila, 12 = Nurmijärvi, 13 = Vieremä 14 =
Kihniö/Parkano, 15 = Iitti, 16 = Imatra/Ruokolahti
One to seven spring cereal fields were examined on each farm, giving a total of 549 fields: 218 under barley
(Hordeum vulgare L.), 183 under oats (Avena sativa L.), 112 under wheat (Triticum aestivum L.), 33 under oat/
pea (Pisum sativum L.) mixture, 2 spring-sown triticale (× Triticosecale Wittmack) and one previously studied
field which was now under maize (Zea mays L., excluded from biomass results). Cover crops, mainly pure Italian
ryegrass (Lolium multiflorum L.) or mixed with white clover (Trifolium repens L.), were sown in 18 fields (20%)
under organic production and 31 fields (7.5%) under conventional production.

Table 1. Number of fields surveyed by region, production system and cereal species
Number of fields
Region Production system Cereal species7
Year/ Municipality No Zone5 Total Conventional6 Organic Barley Oat Wheat

2020
Nauvo/Korppoo 3 S 25 23 (18) 2 6 14 5
Mikkeli 1
7 E 25 17 (17) 8 7 7 4
Lieto/Paimio 2
8 S 44 40 (36) 4 10 13 20
Nivala 10 W 27 21 (20) 6 11 14 2
Nurmijärvi 12 S 41 38 (38) 3 22 7 12
2021
Tammela 4 S 42 31 (31) 11 18 16 7
Kitee 6 E 26 21 (21) 5 12 8 6
Laitila 11 S 41 36 (35) 5 20 15 5
Vieremä 13 W 25 21 (21) 4 14 4 1
Kihniö/Parkano 14 W 26 21 (21) 5 8 18 0
Iitti 15 S 28 24 (22) 4 9 12 7
2022
Jokioinen3 1 S 45 35 (35) 10 14 14 12
Lammi 4
2 S 40 35 (33) 5 9 15 16
Laukaa/Toivakka 5 E 37 30 (29) 7 16 17 3
Laihia 9 W 42 37 (37) 5 27 3 8
Imatra/Ruokolahti 16 E 35 29 (27) 6 15 6 4
Total 549 459 (441) 90 218 183 112
1
= incl. Joroinen, Juva, Pieksämäki, Mikkeli, Pertunmaa; = incl. Lieto, Marttila, Paimio, Tarvasjoki; = incl. Humppila, Jokioinen, Forssa,
2 3

Koski Tl, Loimaa, Somero, Ypäjä; 4 = incl. Hämeenkoski, Kärkölä, Mäntsälä, Pukkila; 5 = key to zone abbreviation: S = South, E = East, W =
West; 6 = number of sprayed fields in parentheses; 7 = added to this 33 fields with cereal/pea mixture, 2 triticale and 1 maize

Altogether 459 study fields were cultivated conventionally and 90 organically. Most of the organic fields (82%)
had been ploughed, whereas in conventional fields reduced tillage (43%) and direct drilling (18%) were more
common. The average area of the study fields was 6.9 ha (range 0.3–51.2 ha) and the area of surveyed fields
was in total 3 786 ha. In the largest fields, the investigated area was restricted to a maximum of 21 ha and in
some fields the surveyed field area was adjusted according to previous surveys. The same 16 regions were
surveyed in two previous studies and altogether 275 out of the 549 surveyed fields were the same as in 2007–
2009 (Salonen et al. 2011).

Permission to investigate weeds in survey fields was asked from farmers only in early July, with the aim that all
cropping measures influencing the weed infestation, including tillage, the level of fertilization and weed control,

53
 J. Salonen et al.

were realized according to their normal practices. Information on cropping measures was recorded by
interviewing the farmers afterwards.

Herbicides had been applied in 441 out of 459 conventionally cultivated fields. Pure MCPA (4-chloro-2-
methylphenoxyacetic acid) was used in 4% of treated fields, products containing phenoxy acid mixtures (18%),
pure sulphonylurea products (33%), sulphonylureas as tank mixture with other active ingredients (33%) and the
rest of fields were treated with various tank mixtures of 2–3 commercial products. Six fields (1%) were treated
with springglyphosate only. Chemical control of wild oat (Avena fatua L.) had been carried out in 39 fields (9%),
in most cases with pinoxaden products.

Weed sampling
The weed survey was carried out during a 3-week period starting in mid-July (weeks 28/29), by which time
spring cereals had reached their heading stage in the majority of survey fields and in most cases around one
month had elapsed since herbicide treatment. The growth stage of cereals varied most in 2021 due to rainy
weather conditions which extended the sowing period and later an exceptional drought hampered the growth
of crops and weeds.

The weed sampling protocol was exactly the same as applied since the survey in 1997–1999 (Salonen et al.
2001a). The occurrence of weeds was assessed from 10 sample quadrats (1.0 m 2) randomly placed in each field.
In advance, the positions of sample quadrats in each field were split into a 10 × 10 cell grid using a random
number calculator. On the spot, each sample grid was placed in the field according to the area and shape of
each field. The distance from the field edge was noted for each quadrat.

Weed density was determined by counting the number of plants or shoots of grass weeds by species in a
rectangular frame measuring 0.1 m2 (25 cm × 40 cm), which was a corner area within a larger quadrat
measuring 1.0 m2 (1.0 m × 1.0 m). The larger quadrat was used only for observations on the total number of
species on a presence/ absence scale. The results presented in the tables and figures derive from data collected
from the small 0.1 m2 quadrats and pooled over the 10 samples in each field. A complete list of the additional
weed species found in the presence/absence observation (1 m 2) is given by survey region in the Appendix as a
supplement to the 40 most frequent species presented in Tables 3 and 5.

In four out of ten small sample quadrats, weeds and cereals were cut at the soil surface. The biomass for
Elymus repens, other weed species in total and for cereal crop were weighed after the samples had been dried
in an air-flow dryer at 40 °C for several days. The air-dry biomass results are presented in kg ha -1. The biomass of
cover crops and undersown grasses and clovers were weighed separately and added to the cereal biomass to
represent the pooled biomass of sown crops in certain calculations. Appraisal of herbicide resistance was based
on visual observation of vigorous growth linked with the information about the applied herbicide.

Nomenclature and data analysis

All weed species found in sampling areas were assessed. Afterwards, some genera/taxa, e.g. Galeopsis spp. and
Lamium spp., were pooled because they could not be unequivocally identified to species level at the seedling
stage.

The plant species nomenclature follows Hämet-Ahti et al. (1998), and the abbreviations of species names are
according to the code system by European and Mediterranean Plant Protection Organization (EPPO) available
on the EPPO website (gd.eppo.int). The full scientific names with attribution are given to the 40 most frequent
species (Tables 3 and 5).

The frequency of occurrence over studied fields (hereafter ‘frequency’) refers to the proportion of fields where
a species was found indicating distribution of weed species. Differences in frequency values between
production systems were tested with Fisher’s Exact Test (Table 4). For each field, the total weed density and
biomass indicating weed infestation level were summed, and the averages and median values of abundance are
presented by 16 survey regions. Differences between the regions were tested with log-transformed values using
the MIXED procedure of SAS 9.2 (SAS Institute Inc., Cary, NC, USA). In the linear mixed model, region,
production systems and their interactions were considered as fixed effects and farm as a random effect. If not
mentioned otherwise, the specific results concerning the weed occurrence in conventional production derive
from the 441 fields treated with herbicides, i.e. the 18 non-sprayed conventional fields were excluded.

54
 Agricultural and Food Science (2023) 32: 51–68

Differences in the weed species diversity between two production systems were determined by calculating
Shannon’s diversity (H’) and evenness (E) indices (Shannon and Weaver 1949) and Simpson’s diversity index
(Simpson 1949 ) based data collected on species number and weed densities in small quadrats (see e.g.
Magurran 2013 for the further information on the indices). The Shannon diversity index values (H’) were
calculated for both production systems as follows: H’ = −∑ I pi ln(pi) where pi is the proportion of total individuals
in the ith species in the fields of production system and the Shannon evenness (E) as follows: E = H’ / ln(S)
where H’ is the Shannon diversity index and S is the total number of species in the production system. The
Simpson index was calculated as follows: D = 1-∑n(n-1)/N(N-1) where n = number of individuals of each species
in the production system and N = total number of individuals of all species in the production system.

Results
Species diversity
In total, 138 weed species were found in the large (1.0 m 2) sampling quadrats and 115 in the small (0.1 m2)
quadrats (see Appendix for the species recorded in large quadrats but not included in the list of 40 most
frequent species). Out of the 138 observed species, 62 were annual and 76 perennial species. The majority (i.e.
98 species) of the observed species occurred in less than 5% of the fields studied and 67 species of them in less
than 1% of fields.

Altogether 95 species were observed in organically cropped fields (N = 90) and 124 species in conventionally
cropped, herbicide-treated fields (N = 441). The majority of the species (59%) was shared with cropping
practices, whereas 43 species (31%) were unique for conventional and 14 species (10%) for organically cropped
fields. Indices describing weed species diversity were very similar between production systems (Table 2).

The average observed species number per field was higher in organically farmed fields, 21 (SD = 4.0, min 12,
max 33), than in conventionally farmed sprayed fields, 12 (SD = 5.0., min 1, max 31).

Table 2. Comparison of weed diversity between production systems using Shannon

index for diversity (H´) and evenness (E) and Simpson’s index for dominance (1/D)
calculated using density values for a total number of observed weed species (S)

Production system
Index Conventional (N=441) Organic (N=90)
Diversity (H´) 2.82 2.87

Evenness (E) 0.62 0.69

Dominance (1/D) 10.98 11.35
Species number (S) 124 95
Frequency of weed species
The occurrence of the 40 most frequent weed species in the small 0.1 m 2 sample quadrats is presented by
region (Tables 3 and 5) and the remaining species observed from the larger 1.0 m 2 quadrats are listed by region
(Appendix). Only nine weed species exceeded the frequency of 50% in large quadrats.

The most frequent weeds were primarily broad-leaved species. Viola arvensis, Stellaria media and
Chenopodium album were the only species with overall frequency above 50% (Table 3), and they were
frequently found both in conventionally and organically cropped fields. Nevertheless, there were evident
differences in the ranking list of predominant weed species and their frequency levels between conventionally
and organically cropped fields (Table 4). S. media was the only weed species judged to be herbicide resistant,
with a low frequency of four sprayed fields.

With a frequency of 42%, Poa annua was the most common grass weed before Elymus repens (40%) (Table 3).
Other grass species were found at much lower frequency; Alopecurus geniculatus (5%), Phleum pratense (1%)
and Poa spp. (1%). Altogether, 26 grass species, including volunteer cereal species, were identified in the
sample quadrats (1.0 m2, see Appendix).

55
 J. Salonen et al.

Highly productive perennial broad-leaved weed species occurred more frequently in organic than in
conventional production; Sonchus arvensis had a frequency of 58% in organic and 28% in conventional fields
and Cirsium arvense 51% and 19%, respectively. Typical weed species in grassland rotations, such as Achillea
millefolium 18% vs.
0.5% and Ranunculus repens 43% vs. 6%, were more frequent in organic than in conventional production.

Table 3. Frequencies (%) of 40 most frequent weed species by regions

Year / Region1 Average
Species / Taxon 2020 2021 2022 2020–2022
3 7 8 10 12 4 6 11 13 14 15 1 2 5 9 16
Achillea millefolium L. 8 4 0 0 5 5 8 0 16 12 4 0 0 3 0 0 3

Alopecurus geniculatus L. 0 8 0 0 0 14 8 0 20 15 0 2 0 14 0 0 5
Artemisia vulgaris L. 4 24 5 0 5 7 4 29 0 8 0 7 8 0 5 11 7
Bidens tripartita L. 4 0 2 0 5 2 0 0 0 0 0 4 3 11 0 0 2
Brassica rapa L. ssp. oleifera (DC.) Metzg. 8 20 2 4 2 5 8 0 4 27 18 0 3 3 0 6 6

Capsella bursa-pastoris (L.) Medik. 44 20 18 19 15 24 42 34 32 15 11 22 23 51 19 26 26

Cerastium fontanum Baumg. 4 0 0 0 7 7 0 0 0 0 0 2 8 0 0 0 2

Chenopodium album L. 92 48 70 93 27 48 54 46 60 23 36 67 38 54 38 54 52
Cirsium arvense (L.) Scop. 16 24 27 33 27 29 4 10 32 15 36 27 20 41 17 26 24
Elymus repens (L.) Gould 12 56 30 37 10 33 58 46 72 73 29 38 25 68 33 49 40

Epilobium L. spp. 0 16 0 4 0 7 8 2 8 0 7 0 0 5 2 6 4

Equisetum arvense L. 40 12 27 4 37 17 4 22 8 8 39 33 13 24 10 29 21
Erysimum cheiranthoides L. 36 40 23 44 15 26 19 15 24 23 21 27 23 38 14 23 25
Fallopia convolvulus (L.) À. Löve 28 24 43 56 34 33 19 20 12 12 32 56 65 32 36 51 36
Fumaria officinalis L. 56 48 39 30 63 19 23 20 16 23 29 56 45 43 26 69 38

Galeopsis L. spp.a 20 36 57 52 27 36 23 22 32 35 25 53 70 70 48 40 42

Galium spurium L.b 92 12 77 22 88 62 8 24 4 23 50 62 45 62 52 46 49

Gnaphalium uliginosum L. 0 36 0 0 0 50 50 29 48 77 7 9 35 68 12 40 28
Lamium L. spp. c
92 28 86 4 63 31 0 37 0 0 54 60 23 35 31 57 40
Lapsana communis L. 72 84 41 4 78 48 35 54 0 23 57 40 75 49 21 89 49

Matricaria matricarioides (Less.) Porter 0 28 0 4 2 19 35 15 48 46 0 13 30 38 19 17 19

Myosotis arvensis (L.) Hill 20 64 30 15 5 26 15 5 8 15 14 24 38 43 36 43 25

Persicaria lapathifolia (L.) Gray 20 28 32 37 10 43 15 27 28 31 21 27 48 49 7 51 30
Plantago major L. 20 32 9 4 12 12 38 17 12 12 18 11 18 35 10 34 18
Poa annua L. 4 56 11 59 34 55 35 37 80 65 36 29 30 76 50 29 42

Polygonum aviculare L. 28 20 27 22 22 38 46 32 36 38 29 51 43 65 36 37 36

Ranunculus repens L. 4 0 5 15 2 12 23 2 24 19 11 11 10 30 10 23 12
Sonchus arvensis L. 60 44 34 41 20 26 8 17 52 50 43 27 33 49 14 43 33
Sonchus asper (L.) Hill 16 0 11 0 2 5 0 12 0 0 11 13 13 0 10 0 6
Spergula arvensis L. 32 36 32 41 12 38 46 27 44 42 29 31 38 57 31 34 35

Stachys palustris L. 0 0 0 0 5 0 0 0 0 0 14 2 8 3 0 3 2

Stellaria media (L.) Vill. 72 68 77 67 49 48 27 39 28 31 43 78 78 68 74 60 58

Taraxacum Weber spp. 8 20 43 0 12 33 31 7 8 4 25 16 8 24 17 23 18
Thlaspi arvense L. 8 12 5 4 12 17 4 2 0 12 4 7 15 11 7 0 8
Trifolium L. spp.d 16 12 7 7 2 14 4 0 28 8 7 4 13 38 2 11 10

Tripleurospermum inodorum (L.) Sch. Bip. 80 12 34 26 39 36 4 44 0 15 39 47 35 35 21 40 33

Tussilago farfara L. 0 4 0 0 0 2 8 0 12 4 4 2 0 8 0 9 3
Veronica L. spp.e 8 16 5 0 44 24 12 41 0 0 36 4 28 3 2 29 17
Vicia L. spp.f 24 12 7 15 7 21 12 15 12 27 11 13 18 11 10 20 14

56
 Agricultural and Food Science (2023) 32: 51–68

Viola arvensis Murrayg 80 100 59 74 51 69 81 76 84 62 46 78 90 89 88 86 75

1
Key to regions given in Figure 1 and Table 1. a= mainly G. bifida and G. speciosa; b= incl. G. aparine; c= mainly L. purpureum; d= mainly T.
repens; e= mainly V. agrestis; f=mainly V. cracca; g=incl. V. tricolor
Table 4. Frequency of ten most common weed species in two production systems and the change in frequency-% from 2007–
2009 to 2020–2022
Sprayed conventional (N = 441) Organic (N = 90)

Rank Weed species % Change Weed species % Change

1 Viola arvensis 73 –10 Chenopodium album 99* +3
2 Stellaria media 52 –13 Erysimum cheiranthoides 84* –2
3 Galium spurium 50 –9 Viola arvensis 83* –11
4 Poa annua 48 +15 Stellaria media 82* –12
5 Lapsana communis 47 –10 Galeopsis spp. 81* –7
6 Chenopodium album 41 –11 Elymus repens 79* –10
7 Lamium spp. 39 –3 Spergula arvensis 74* –15
8 Fumaria officinalis 37 –11 Persicaria lapathifolia 59* –9
9 Galeopsis spp. 33 –26 Vicia spp. 59* +9
10 Polygonum aviculare 33 –15 Sonchus arvensis 58* –3

Erysimum cheiranthoides# 12 –13 Galium spurium# 40 +1

Elymus repens 32 –18 Lapsana communis 56 –4

Spergula arvensis 26 –12 Poa annua 11* –4

Persicaria lapathifolia 24 –6 Lamium spp. 41 +16

Vicia spp. 5 –4 Fumaria officinalis 49* –9

Sonchus arvensis 28 –7 Polygonum aviculare 50* –22

* = indicates significant difference in species frequencies between production systems (’Fisher’s Exact Test, p<0.05); # = Cross-comparison
to the species in another production system

Some weed species were characteristic of certain regions in Finland; Galium spurium, Sonchus asper,
Tripleurospermum inodorum and Veronica spp. were most frequently found in southern survey regions,
whereas the frequency of Elymus repens, Gnaphalium uliginosum, Matricaria matricarioides, and Poa annua
increased towards the east and north. Two common species, Lamium spp. and Lapsana communis, were
missing or rare in some regions (6, 10, 13, 14).

The number of fields with observed wild oat infestation was 36 in total (6.5% of all fields), all of them under
conventional production. In the previous survey 2007–2009 the proportion of infested fields was almost 30%,
but this figure was based on the information from farmers and therefore direct comparison is not possible.

Weed density
The average density of weeds in all fields surveyed was 189 plants m-2 (SD = 196, median = 123, N = 549) (Table
5). However, the difference between the two production systems was considerable; the average density of
weeds in organic production was 384 plants m-2 (SD = 248, median = 347, N = 90), in sprayed conventional fields
147 plants m-2 (SD = 153, median = 97, N = 441) and in unsprayed conventional fields 261 plants m-2 (SD = 251,
median = 177, N = 18). The total weed densities were lowest in conventional production in the regions 10 (74
plants m-2) and 15 (86 plants m-2), differing significantly (p < 0.05) from the regions 1, 2, 3, 5, 7, 8, 13 and 16.
Regional differences in organic fields were not found.

Ranking of the most abundant species differed between the cropping systems both in terms of species
composition and the level of weed density (Fig. 2). The three most abundant weed species in organic fields

57
 J. Salonen et al.

were Chenopodium album (on average 73 plants m-2 vs. 10 in sprayed conventional fields), Spergula arvensis (50
vs. 3) and Elymus repens (48 vs. 7). Correspondingly, the three most abundant weed species in sprayed
conventional fields were Poa annua (25 vs. 1), Viola arvensis (24 vs. 19) and Stellaria media (18 vs. 26). In
conventional fields, the average density of E. repens had a decreasing trend from 12 plants m -2 in 2007–2009 to
7 plants m-2 and likewise in organic fields from 68 plants m-2 to 48 plants m-2.

58
 Table 5. Weed density (plants m-2) by regions. + indicates Mean < 1 plant m-2

1
Average
Species / Taxon Year / Region
2020–2022

2020 2021 2022

8 12 13 15 5 16
3 7 10 4 6 11 14 1 2 9
Achillea millefo/ium L. + 1 0 0 + + 1 0 + 1 + 0 0 + 0 0 +
Alopecurus geniculatus L. 0 + 0 0 0 1 + 0 1 + 0 + 0 + 0 0 1
Artemisia vulgaris L. + + + 0 + + + 1 0 + 0 + + 0 + 1 +
Bidens tripartita L. + 0 + 0 + + 0 0 0 0 0 + + + 0 0 +
Brassica rapa L. ssp. oleifera (DC.) Metzg + 1 + + + + + 0 3 2 2 0 + + 0 + +

Capsella bursa-pastaris (L.) Medik. 6 2 2 + 1 3 2 2 2 1 1 1 + 2 + 3 2 J. Salonen et al

Cerastium fontanum Baumg. + 0 0 0 + + 0 0 0 0 0 + + 0 0 0 +
58
Chenopodium album L. 56 41 11 17 6 19 58 9 8 3 3 76 30 16 5 13 23
Cirsium arvense (L.) Scop. + 1 1 1 1 1 + 1 2 1 1 1 + 1 1 1 1
Elymus repens (L.) Gould 6 19 19 20 2 12 21 11 11 25 7 9 3 26 2 37 14

Epilobium L. spp. 0 + 0 + 0 + + + + 0 + 0 0 + + + +
Equisetum arvense L. 2 1 1 + 2 + + 1 + + 2 2 + + + 1 1
Erysimum cheiranthoides L. 2 3 1 5 2 1 + 2 3 4 2 3 4 4 1 1 2
Fallopia convolvulus (L.) À. Löve 1 1 5 2 1 2 2 + + + + 3 8 4 3 4 3
Fumaria officinalis L. 4 5 2 2 9 1 + + + 1 1 8 4 2 4 13 4

Galeopsis L. spp.a 2 13 4 2 2 11 4 1 1 1 2 6 22 4 10 3 6
Galium spurium L.b 38 2 14 1 30 10 + 2 + 4 3 11 6 3 5 4 9
 Gnaphalium uliginosum L. 0 5 0 0 0 17 15 8 9 25 + + 8 24 7 3 7
Lamium L. spp. c
34 2 19 + 12 2 0 5 0 0 17 26 2 2 7 10 9
Lapsana communis L. 7 12 2 + 23 4 3 9 0 1 5 3 18 7 + 15 7
Table 5 continues

Matricaria matricarioides (Less.) Porter 0 7 0 + + 2 2 1 5 3 0 + 2 7 3 4 2

Myosotis arvensis (L.) Hill + 9 + 1 + 3 1 + + + + 1 2 4 3 3 2
Persicaria lapathifolia (L.) Gray 2 1 3 1 1 3 6 3 4 3 5 7 5 6 + 6 3
Plantago major L. 1 3 + + 1 2 4 1 2 + 1 1 1 6 3 5 2
Poa annua L. + 50 7 10 10 16 9 10 84 27 13 18 14 61 17 8 21
Agricultural and

Polygonum aviculare L. 1 + + 1 1 1 1 1 1 1 1 2 2 3 2 1 1
Ranunculus repens L. + 0 + 1 2 1 + + 2 2 + + + 4 + 2 1
Sonchus arvensis L. 3 2 2 5 1 3 1 2 4 5 3 1 2 4 + 11 3
Sonchus asper (L.) Hill + 0 1 0 1 1 0 + 0 0 + + 2 0 1 0 +
Spergula arvensis L. 6 15 9 57 2 6 9 3 13 37 17 3 7 12 1 10 11
59

Stachys palustris L. 0 0 0 0 + 0 0 0 0 0 1 + + + 0 + +
Stellaria media (L.) Vill. 35 14 55 11 15 5 2 2 2 1 13 44 40 12 32 5 20
Taraxacum Weber spp. 1 1 1 0 + 1 1 + + + 1 + + 1 + 1 1
Thlaspi arvense L. 1 2 + + 1 1 + + 0 + + + 1 + 2 0 +
Trifolium L. spp. d
2 + + + + 1 3 0 3 + + + 1 2 + + 1

Tripleurospermum inodorum (L.) Sch. Bip. 6 1 4 2 5 3 + 2 0 + 1 5 2 2 1 7 3

Tussilago farfara L. 0 + 0 0 0 + + 0 1 + + + 0 + 0 1 +
Veronica L. spp.e + 3 + 0 20 1 + 9 0 0 9 1 9 1 + 4 4
Vicia L. spp.f 1 + 1 2 1 1 + 2 1 1 1 1 1 1 + 1 1
Viola arvensis Murray g
7 33 5 12 12 18 19 27 16 32 6 24 66 25 34 25 23
Mean total, plants m-2 228 251 171 153 163 154 167 121 178 193 122 260 263 263 153 205 189
Median total, plants m -2
175 209 111 64 101 96 84 87 123 131 76 128 231 225 113 153 123

1
Key to regions given in Figure 1 and Table 1. a= mainly G. bifida and G. speciosa, b= incl. G. aparine, c= mainly L. purpureum, d= mainly T. repens, e= mainly V. agrestis, f=mainly V. cracca, g=incl. V.
tricolor
 J. Salonen et al.

Sprayed conventional (N=441)

POLLA
POLCO
FUMOF
SPRAR
GAESS
VERSS
GNAUL
AGRRE
LAMSS
LAPCO
GALSP
CHEAL
STEME
VIOAR
POAAN

0 5 10 15 20 25 30

Plants m-2

Organic (N=90)
LAPCO
MATMT
FUMOF
MATIN
POLLA
SONAR
ERYCH
GNAUL
LAMSS
VIOAR
GAESS
STEME
AGRRE
SPRAR
CHEAL
0 10 20 30 40 50 60 70 80
-2
Plants m

Fig. 2. Average weed density (plants m -2) of most abundant weed species in sprayed
conventional and organically grown spring cereal fields. EPPO codes for weed species available
at https:// gd.eppo.int/. Note different scale on x-axis.

Weed biomass
The average biomass production of weeds was 246 kg ha-1 (SD = 377, median = 98, N = 547) (Table 6). The
difference between production systems was as clear as in the case of weed densities, in that for organic fields
the average weed biomass was 678 kg ha -1 (SD = 568, median = 538, N = 90), being more than four times higher
than the quantity of 151 kg ha-1 (SD = 243, median = 70, N = 440) grown in sprayed conventional fields. In
unsprayed conventional fields the corresponding value was 399 kg ha -1 (SD = 327, median = 338, N = 17). The
deviant total biomass in the region 7 (593 kg ha -1) was due to substantially high biomass of Chenopodium album
some organic fields and high biomass of Poa annua in both production systems.
Table 6. Weed biomass (kg ha-1) by regions.
Year / Region1 Average
2020 2021 2022 2020–2022
Category
3 7 8 10 12 4 6 11 13 14 15 1 2 5 9 16
Elymus repens (L.) Gould 22 123 36 82 6 49 79 44 103 151 8 27 10 101 15 98 54

Other weeds in total 258 471 126 100 134 209 268 196 234 232 102 154 163 184 137 246 191
Mean total, kg ha-1 280 593 162 182 141 258 347 240 337 383 110 181 173 285 152 344 246
Median total, kg ha-1 163 368 38 90 53 74 105 98 158 224 49 64 74 173 78 168 98

1 ) Key to regions given in Figure 1 and Table 1

62
 Agricultural and Food Science (2023) 32: 51–68

There was a marked difference in the air-dried biomass of cereal crops between production systems, being on
average 3 180 kg ha-1 (median = 3 139) in organic fields and 5 675 kg ha -1 (median = 5 639) in sprayed
conventional fields. The proportion of weed biomass relative to total vegetative biomass (crop + weeds) was
relatively low (mean = 2.7%) in sprayed conventional fields, but considerably higher (mean = 21%) in organic
fields (Fig. 3). Furthermore, the proportion of sprayed fields remained high, 96% of studied conventional fields,
and the proportion of weeds out of the total biomass of crop stand was below 5% in 83% of sprayed fields,
whereas it was more than 10% in 68% of organic fields.

Fields (%)
Conventional Organic
50

40

30

20

10

0
<1% 1-5% 5-10% 10-25% >25%

Proportion of weed biomass,%

Fig 3. Relative number of fields (%) in different weed biomass classes in two production systems.
Weed biomass is proportioned to total vegetative biomass (crop + weeds).

The average biomass of Elymus repens in organic fields was 160 kg ha -1, accounting for 24% of the total weed
biomass pooled over all organic fields and correspondingly the average biomass 34 kg ha -1 accounted for 22% in
sprayed conventional fields.

Discussion
We found only minor changes in weed species diversity compared with the previous weed survey in 2007–2009
(Salonen et al. 2011). However, changes in the total occurrence of single species were dependent on production
system.

The number of weed species, 115, found in small sample quadrats (0.1 m 2), was slightly lower than in previous
survey when 148 species were recorded (Salonen et al. 2011). In the larger (1.0 m 2) quadrats, the number of
observed species was now 138 versus 175 species, respectively. However, the average number of weed species
per field remained the same in both production systems compared with the previous survey. The most
probable reason for the reduction in species diversity is that the total number of surveyed fields was lower (549
vs. 595) (cf. Pollnac et al. 2009) but the decline can also be regarded as a consequence of agricultural
intensification (Cirujeda et al. 2011, Richner et al. 2015, Chamorro et al. 2016, Andreasen et al. 2018). In
comparison, Mukula et al. (1969) recorded 304 weed species in the first national weed survey in Finland, and
188 species were recorded in the larger quadrats of our weed survey in the 1990s (Salonen et al. 2001a).

The total number of weed species was found to be higher in conventional than in organic fields, a similar finding
with the earlier weed survey (Salonen et al. 2011). This was likely due to five times higher sampling effort of
conventional than organic fields. Despite of differences in the total weed species number between production

63
 J. Salonen et al.

systems, the differences in diversity and evenness indices were minor. A similar result has earlier been found in
the comparisons of weed species diversity between production systems (Hyvönen and Salonen 2002).

Only 34 weed species exceeded the overall frequency level of 5%. Moreover, the list of 40 most frequent
species was dominated by broad-leaved species and only three grass species were included. Although we found
more perennial than annual species, only 16 perennial species out of 76 recorded ones were included in the list
of 40 most frequent species.

In all, the frequencies of occurrence for the most common weed species were significantly higher in organic
cropping than in conventional cropping. Characteristic species of organic production, Chenopodium album,
Erysimum cheiranthoides and Spergula arvensis, were effectively controlled with herbicides in conventional
production. Viola arvensis continues to be a characteristic species of cereal fields, as reported in other surveys
(Andreasen and Stryhn 2008, Májeková et al. 2019, Hofmeijer et al. 2021). Only four weed species out of ten
were the same in the list of the most frequent species in both production systems. Competitive perennial
species Cirsium arvense and Sonchus arvensis were especially associated with organic production both in terms
of frequency and density.

Concerning apparent changes among the common weed species, Poa annua, (+11% units increase since 2007–
2009), and Veronica spp. (+8), have become more frequent, whereas Myosotis arvensis (–23), Taraxacum spp.
(–21), Polygonum aviculare (–16) and Elymus repens (–16) have become less frequent since 2007–2009
(Salonen et al. 2011). There were three newcomers in the list of 40 most frequent species: Bidens tripartita,
Stachys palustris and Tussilago farfara, which replaced Juncus bufonius L., Phleum pratense L. and Poa
pratensis L.

The increased occurrence of Poa annua is noteworthy because its frequency (42%) significantly increased since
the two previous surveys, having been 31% in 2007–2009 (Salonen et al. 2011) and only 12% in 1997–1999
(Salonen et al. 2001a). However, P. annua was characteristic only in conventional production in terms of
frequency (48%) and average density (25 plants m -2) compared with organic production (11% and 1 plant m -2).
The success of P. annua is in line with observations in the Nordic and Baltic countries (Tørresen et al. 2006,
Andreasen and Stryhn 2008, Nečajeva et al. 2018) linked with moist soil conditions, adaptation to compact soil,
successful overwintering and substantial phenotypic and genotypic variability (Warwick 1979). Increased area
of reduced tillage in survey fields since the 1990s has likely favoured P. annua, as reported earlier in Norway
(Tørresen et al. 2003).

The abundance of weeds remained at the same level in conventional but declined in organic production since
the previous survey in 2007–2009 (Salonen et al. 2011). The decline in total weed biomass in organic
production was primarily due to Elymus repens, with 34% lower biomass than in 2007–2009. E. repens is still
the most harmful weed species for spring cereals in Finland, producing on average 23% of total weed biomass
in both production systems compared with an average of 29% in the earlier survey. In conventional production,
the declining trend of E. repens biomass began in the late 1990s (Salonen et al. 2013) as a consequence of
common use of glyphosate, which has been the most sold herbicide in Finland for more than two decades
(Tukes 2023). In comparison, a similar decline in E. repens was observed earlier in Denmark (Andreasen and
Stryhn 2008) and Latvia (Nečajeva et al. 2018), and also associated with increased use of glyphosate.

The benefit of using herbicides in reducing weed competition against crop is clearly demonstrated by
comparing weed biomass results from sprayed and unsprayed conventional fields. Weed management in
conventional production of spring cereals has relied on herbicide use in Finland, as well as in 96% of
conventional fields in this survey. This has resulted in relatively low total abundance of weeds and a declining
trend in Elymus repens infestation. The scenario for the future is strongly dependent on the registration of
glyphosate (EC 2022) and the actions to reduce the overall use and risk of chemical pesticides by 50% by 2030
in the EU (EC 2020).

As a newly observed challenge, increased infestation of Poa annua calls for rapid development of weed
management strategies in conventional production. In contrast, the evolving risk of herbicide resistance in
Stellaria media has been acknowledged and successfully taken into account among farmers who have adapted
their herbicide use either by replacing or supplementing sulfonylureas with products with different modes of
action.

64
 Agricultural and Food Science (2023) 32: 51–68

Conclusions
As demonstrated in our survey, weed communities provide floral diversity to spring cereal fields but also
represent a notable threat to cereal production not only in organic production but also in unsprayed
conventional fields. Integrated cropping measures that promote crop competition as part of an integrated weed
management strategy will become necessary to reach EU pesticide targets. Efforts to establish competitive crop
stands and develop mechanical weed control methods are important for both production systems.

Acknowledgements
The survey was a follow-up study in the series of weed monitoring in spring cereal fields started in the early
1960s, currently financed by Luke’s statutory and expert services. We thank all farmers who allowed us to visit
their fields and provided us with information about their farming practices. Our permanent technical staff,
Leena Ruokonen, Susanna Lehtonen and Sanna Maula and a team of field assistants Katja Kauppi, Juha
Nykänen, Leo Laaksonen and Kukka Kutilainen, are acknowledged for their efforts in field work and data
processing.

Appendix
Frequency (% in descending order in each region) of additional weed species observed in survey fields (N=549) by
region. EPPO codes for weed species available at https://gd.eppo.int/

1. Jokioinen (45 fields) 2. Lammi (40) 3. Nauvo/Korppoo(25) 4. Tammela (42)

Weed Freq-% Weed Freq-% Weed Freq-% Weed Freq-%
LTHPR 4 POLHY 5 EPHHE 36 IUNBU 14
BARVU 2 SECCE 5 CENCY 8 POLHY 14
CIRSS 2 ALOPR 3 GERSY 8 PHLPR 12
CRYCA 2 BARVU 3 SENVU 8 RORSY 12
EPHHE 2 LEBAU 3 ACHPT 4 AVESA 7
PHLPR 2 PHLPR 3 CHYLE 4 BARVU 7
PHRCO 2 PIEAB 3 LINVU 4 HORVS 7
POLPE 2 POAPR 3 MENAR 4 SENVU 7
RORSY 2 PTLAN 3 RAPRA 4 AGSTE 5
RUMSS 2 RUMSS 3 SONOL 4 LTHPR 5
URTDI 2 SENVU 3 RUMAA 5

SALCA 5

VIOSS 5

ACHPT 2

ANKSY 2

ANRSY 2

ATXPA 2

AVEFA 2

CIRSS 2

DACGL 2

65
 J. Salonen et al.

FAGES 2

MEDSS 2

POAPR 2

POATR 2

RUBID 2

RUMSS 2

TRZAS 2

URTDI 2

5. Laukaa/Toivakka (37) 6. Kitee (26) 7. Mikkeli (25) 8. Lieto/Paimio (44)

Weed Freq-% Weed Freq-% Weed Freq-% Weed Freq-%
RAPRA 16 RORSY 15 CRYCA 8 BRSNN 11
RUMSS 14 URTDI 15 LINVU 8 RUMSS 7
CRYCA 8 LINVU 12 AEOPO 4 PHLPR 7
CIRSS 5 LTHPR 12 APESS 4 PIUSI 5
DECCA 5 ACHPT 8 CENCY 4 CIRSS 2
EPICT 5 AVESA 8 HORVS 4 EPHHE 2
PRUVU 5 PHLPR 8 LTHPR 4 FESPR 2
URTUR 5 RUMAA 8 PHLPR 4 LOLSS 2
FAGES 3 ALCSS 4 POATR 4 LTHPR 2
LEBAU 3 HORVS 4 POLHY 4 PIEAB 2
PHLPR 3 IUNBU 4 SENVU 2

PHRCO 3 KNAAR 4 VICFX 2

PIBSA 3 LEBAU 4

PIUSI 3 POLHY 4

POAPR 3 RAPRA 4

POLHY 3 TRZAS 4

RORPA 3 VIOSS 4

RUMAA 3

SENVU 3

SINAL 3

URTDI 3

9. Laihia (42) 10. Nivala (27) 11. Laitila (41) 12. Nurmijärvi (41)
Weed Freq-% Weed Freq-% Weed Freq-% Weed Freq-%
AVESA 14 HORVS 26 CRYCA 10
AVEFA 2
PIEAB 7 LEBAU 11 CHYLE 2 LOLSS 7
POLPE 7 RUMSS 7 EPHHE 2 LEBAU 5
HORVS 7 RUMAA 7 EPIAC 2 PHRCO 5

66
 Agricultural and Food Science (2023) 32: 51–68

MEDSS 7 RORSY 4 PHLPR 2 POLPE 5

CVPTE 5 CENCY 4 SOLTU 2 SENVU 5
PHLPR 5 LTHPR 4 SONOL 2 ACHPT 2
PIUSI 5 POAPR 4 ANKSY 2

APESS 2 RORIS 4 AVESA 2

CIRSS 2 RUMAC 4 BARVU 2

LEBAU 2 URTDI 4 BRSNN 2

RORSY 2 VERLO 4 CIRSS 2

VIOSS 4 CLMAR 2

FESPR 2

LINVU 2

LTHPR 2

PHLPR 2

PIEAB 2

PIUSI 2

POLHY 2

RUMSS 2

SONOL 2

URTDI 2

13. Vieremä (25) 14. Kihniö/Parkano (26) 15. Iitti (28) 16. Imatra/Ruoholahti (35)
Weed Freq-% Weed Freq-% Weed Freq-% Weed Freq-%
AVESA 24 HORVS 23 PTLAN 11 LINVU 11
LTHPR 16 CHYLE 15 EROCI 7 PHCTA 9
LEBAU 12 AVESA 12 LEBAU 7 CHYLE 6
RUMAA 12 POLHY 12 RUMAA 7 EROCI 6
HORVS 8 RANAC 12 TTTMM 7 PHLPR 6
PHLPR 8 ACHPT 8 ACHPT 4 PTLAN 6
PIUSI 8 FESPR 8 CRYCA 4 URTDI 6
RUMSS 8 LOLPE 8 LINVU 4 AEOPO 3
FESOV 4 LTHPR 8 LTHPR 4 CNISA 3
HIEJU 4 RUMAA 8 POLHY 4 HYPMA 3
LINVU 4 IUNBU 4 SAXSS 4 MENAR 3
LUPPO 4 LEBAU 4 PHRCO 3

PHCTA 4 MELRU 4 POAPR 3

PTLER 4 RUMSS 4 POATR 3

RORIS 4 SAIPR 4 RUBID 3

67
 J. Salonen et al.

URTDI 4 TRZAS 3

68