1. Introduction
Fallopia Japonica (F. Japonica) is a fast-spreading invasive herbaceous plant mainly known for its ecosystem disservices.
F. japonica causes considerable economic and environmental damage throughout the USA, UK, and Europe [
1,
2]. Due to its ability to quickly spread, it reduces the diversity [
3,
4,
5] and the activity [
6] of native biota, increases soil erodibility [
7], affects temporal patterns of soil nutrient availability [
8], and causes significant structural and functional changes in urban and rural ecosystems [
9]. As such
F. japonica poses considerable economical, planning, and logistical problem in urban and rural land management [
10].
To control its spread, different mechanical, chemical, and biological approaches have been applied individually or in combination. Control methods that integrate a variety of options can lead to near-effective control of
F. japonica [
11]. Mechanical control methods, such as hand cutting, mowing, grazing, digging, pulling, and covering have proven labor-intensive but in some cases still lackluster in limiting
F. japonica’s growth. The spread of
F. japonica is to some extent controllable with herbicides [
12], and biological agents, such as psyllid
Aphalara itadori Shinji [
13] and the fungus
Mycosphaerella polygoni-cuspidati [
14]. The restoration of contaminated areas requires careful species selection due to the allelopathic effects of
F. japonica [
15]. For example, fast-growing
Salix viminalis significantly reduces knotweed spread due to its rapid growth and, therefore, represents promising restoration of species in some urban environments and riparian zones [
16].
F. japonica is also known for its numerous ecosystem services. The aerial or underground parts of
F. japonica show insecticidal [
17] and fungicidal potential [
18], and have been successfully used in the production of medicine [
19], biofuel [
20], briquettes for heating [
21], paper cellulose fibers [
22], textile [
23], and as carbon (C) adsorbent [
24]. In some of these production processes, the aerial parts of the plant went unused and were disposed of in landfills, although these parts weigh more than half the total mass of the plants [
12,
19]. To avoid resource waste and develop new potential benefits, it was suggested to focus research and development on the added value of the aerial parts [
19]. Existing research has not yet focused on the potential uses of the aerial parts on
F. japonica in the production of organic fertilizer (OF), although the high nitrogen (N) sequestering and remobilization ability of
F. japonica [
1] indicates that such uses might be possible, and attempts in composting
F. japonica for the production of OF have proven promising [
25].
We tested the aerial parts (leaves and stems) of
F. japonica as raw material for the production of OF and applied the fertilizer on a field to evaluate
F. japonica’s effects on Chinese cabbage and agricultural soil. OF was, for the first time, produced by a fermentation process using a microbial inoculant known as effective microorganisms (EM). It is reported that fermented OF possesses advantages over composts, including easier and more environmentally friendly preparation from the raw material [
26]. The use of fermented products rich in selected microorganisms has positive effects on soil biological activity and may improve physical and chemical soil properties, and plant growth [
27,
28,
29].
We hypothesized that the OF produced from F. japonica would successfully replace farmyard manure, which is not easily obtainable or handled in an urban environment. Furthermore, we assumed that mineralization and nutrient availability would be faster than those from other, comparable composts. This approach could serve as an alternative method for managing the uncontrolled spread of F. japonica in urban areas.
2. Materials and Methods
The presented experiment focuses on the management problem of
F. japonica in the case-study city of Ljubljana (Slovenia), where approximately 5 ha of municipality‒owned land is infected by
F. japonica stands, and a further 30 ha is infested by
F. × bohemica [
30,
31] (at least 0.13% of the total municipal area) (
Figure 1a). The experiment was conducted in August, September, and October 2017. The average monthly air temperature in 2017 was 11.9 °C, and the annual precipitation was 1531 mm. The monthly average temperatures were about 1 °C above, 1 °C below, and within the long-term average (1981–2010) during all three months. The monthly precipitation in August and October was half the long-term average. September was extremely wet, with twice the amount of the long-term average precipitation. The daily maximum temperatures in months 8, 9, and 10 were 30.2 °C, 19.1 °C, and 18.7 °C, respectively. The minimum daily temperatures in months 8, 9, and 10 were 16.7 °C, 11 °C, and 6.7 °C, respectively [
32]. Although the heatwave shortly after planting in early August slowed down the initial growth of the plants, the conditions in September and October were favorable for the growth of Chinese cabbage (
Figure 1b).
The aerial parts of F. japonica were harvested from infested grassland near a recreational pathway (46°02′48.36″ N, 14°33′51.80″ E) on 19th May 2017. Each shoot was cut approximately 5 cm above the ground to avoid harvesting roots with high vegetative reproduction potential. The plant material was collected from an unmanaged patch (29% of the harvested area) and from a one‒month‒old patch (71% of the harvested area). In total, 100 kg of fresh plant material was collected from a total area of 42 m2.
The plant material was stored in jute bags and transported to the facilities of the Biotechnical Faculty of the University of Ljubljana, where it was air dried at 40 °C for two days and then cut by machine in to pieces 1–5 cm long. The chopped plant material was transferred into two plastic 73- and 62-L barrels with screw caps, which were filled up to 57 L (i.e., three‒quarters of a barrel) and 31 L (half a barrel), respectively. The plant material was layered into the barrel in approximately 2 cm thick horizons. One fistful (30 g) of EM Bokashi—a commercial product made of lactic acid bacteria, yeast, and photosynthetic bacteria grown on a substrate of cereal bran and molasses—was spread on each horizon (30 g per 5.6 L or ca. 850 g of freshly chopped F. japonica = 35 g/kg). The mass was then firmly pressed, expelling air from the pores.
The plant material was covered with plastic foil, weighted with bricks and plastic bags filled with water, and closed with the screw caps to ensure no air entered the barrels during the process. The material was left to ferment in a dark and unheated room for two months. The average air temperature during fermentation was around 20 °C during the day and 10 °C during the night. The input plant material before fermentation and the output OF after fermentation were analyzed according to the Weender analysis—a partitioning and quantification of moisture (103 °C, 4 h), crude ash (ISO 5984:2002), crude protein (ISO 59832:2009) (or Kjeldahl protein), crude oils and fats (EC 152/2009, annex II H Procedure A), crude fiber (ISO 6865:2000), and N-free extracts (digestible carbohydrates). The fermented
F. japonica to be further used in the experiment as OF was analyzed for organic fatty acids (water extraction, followed by esterification and analysis with GC-FID) (
Table 1 and
Table 2). A composite (bulk) sample of 15 individual samples (increments) was collected for both analyses (ISO 18400-102:2018).
The fertilization experiment was carried out in the multipurpose public urban green area, LivadaLAB (46°01′47.95″ N, 14°30′37.54″ E). The soil of the experimental field was a typical mineralized organic soil over a former lake and river sediment of lime gyttja and clay, and the texture was silt-loam. The characteristics of the plough horizon (0–20 cm) were as follows: pH (CaCl
2) 6.5, plant available phosphorus (P
2O
5) 45 mg kg
−1 dry soil (ammonium lactate method; A level according to the Slovene guidelines for professional fertilization [
33]), and plant available potassium (K
2O) 122 mg kg
−1 soil (ammonium lactate method; B level [
33]). The soil contained 12.55% of organic matter, 7.25% C, 0.64% N, and a C-to-N ratio of 11:5. In the last 10 years, the soil at the site was not utilized for agricultural production, not fertilized, and used as grassland, and the vegetation was cut once a year to prevent overgrowing with bushes and trees. The cut grass was removed from the site.
The soil was prepared by plowing with a moldboard plow 25 cm deep. Next, we crumbled the furrows and prepared the soil by harrowing, leveling, and forming raised beds. The soil was then saturated with water up to approximately 80% of the field’s capacity, which was suitable for transplanting the seedlings of Chinese cabbage.
The planting date was August 6, 2017, and the harvest date was 17 October 2017. The seedlings with four true leaves (approx. 8 cm in height) of Chinese cabbage were transplanted into the soil at a density of 14 Yuki F1 plants per 1 m
2 at a distance of 30 cm in each row, with 20 cm between rows. The crop was treated with OF at a rate of 0, 1, and 2 kg m
−2; the total amount of N applied in each treatment was 0 kg N ha
−1 (treatment N0), 159 kg N ha
−1 (treatment N159; small dose; 10 t ha
−1), and 317 kg N ha
−1 (treatment N317, large dose; 20 t ha
−1) (
Table 1). The OF was incorporated in soil (5 cm deep) manually just before planting.
The total area of the experimental site was 22.14 m2 (5.4. m × 4.1 m), organized as three slightly raised beds (ca. 10 cm high). Three experimental blocks were prepared, and one replicate of each treatment was randomly assigned in each block. Protection bands 0.5 m wide were formed between experimental blocks and bands 0.3 m wide were formed between experimental plots within each block to prevent the influence of fertilizers from the neighboring sectors. Each plot had an area of 1 m2 (1.4 m × 0.7 m).
Nearly all of the N in the OF was organically bound. Only traces were found in the form of ammonium, which is surprising, as anaerobic digestion usually leads to a greater amount of ammonia [
34,
35]. Nitrate and nitrite forms were measured but not detected (
Table 2). The chosen fertilization rates fell both below and above the common practice of vegetable growers, who use a fertilization target value of 240 kg ha
−1 mineral N to achieve 50 t ha
−1 of fresh Chinese cabbage yield [
33]. Since all the N was organically bound in the OF, we expected that only a small percentage would be released over the two months of our Chinese cabbage growth experiment.
Pesticides were not used in the experiment. The plants were irrigated according to visual observations every second day in August using an irrigation bucket. Then, the plants were irrigated every fourth day. We irrigated approximately 10 mm of water per irrigation to the point that the first layer of soil (up to 4 cm) was determined to be wet with a finger test.
The yield of standard and non-standard heads and the total masses of plants were then measured by weighing. Five random plants from each plot were selected; one quarter from each plant was taken for further analysis (ISO 24153:2009). Plant samples were air-dried in a fan-aerated chamber at 40 °C to a constant weight and milled with a Retsch Cutting Mill SM 100 to a final fineness smaller than 1.0 mm. The air-dried and milled bulk samples were further analyzed for their total dry matter (SIST EN 13040:2008). The analyses for total C and N were performed with a Vario MAX instrument (Elementar Analysensysteme GmbH). This instrument operates on the principle of dry combustion, by simultaneously analyzing the C and N content and capturing organic and inorganic forms (SIST ISO 10694; SIST ISO 13878).
Ammonium and nitrate content were extracted with 0.01 M CaCl
2 through 2 h of shaking. Then, we filtered the extract through a 0.45 µm filter and measured ammonium and nitrate using a Thermo Scientific ™ Gallery ™ Plus Automated Photometric Analyzer (SIST ISO 14255:1999). For the elemental analyses of total P and K content, the samples were digested with multiple acids, which dissolved most of the minerals; 0.25 g of each sample was heated first in HNO
3, then in HClO
4, and lastly in HF until evaporation. Each sample was then dried, the residue was dissolved in HCl, and the elements were measured using an ICP-ES Inductively Coupled Plasma Emission Spectrometer [
36].
Program R was used for statistical analyses, and the differences between treatments were estimated with the least significant difference test. The significance level was set at α = 0.05.
Finally, the potential impact on urban food production was estimated based on the average yield of the fresh mass of
F. japonica from the vegetation canopy in July (90 t ha
−1) [
37]. We modeled the required application of fertilizer from
F. japonica needed to cover the plant nutrient requirements of the selected vegetable crops at a given expected yield. The selection of crops was based on the research of Glavan et al. [
38], who provided a detailed list of the 14 most commonly grown crops by urban allotment farmers in Ljubljana. This list represents the average allotment for a garden in Ljubljana. The calculation is based on the expected yield (t ha
−1) and expected crop nutrient uptake (kg ha
−1) given in the technological instructions for integrated vegetable production—plant production based on integrated plant nutrient and pest management [
39].
4. Discussion
F. japonica is known for its participation in numerous ecosystem services [
19,
20,
21,
22]. However, as an invasive species, it causes significant structural and functional changes in urban and rural ecosystems [
1,
2,
3,
4,
5,
8,
9,
10]. This research focused on the added value of
F. japonica [
19]. The rationale for this research was provided by Aguilera et al. [
1], who indicated a high N uptake from the soil for
F. japonica, which led us to assume the good N remobilization capacity of the produced organic fertilizer (OF) from
F. japonica.
The fresh plant biomass harvested in this experiment, 23.8 t ha
−1 was significantly lower than the 47 t ha
−1 reported by Bernik et al. [
37], who reported a harvest from an unmanaged patch. However, the plant material in this research was collected from a combination of unmanaged and managed one-month-old patches, which is why the green mass was less developed and less fresh plant biomass was harvested.
A fermentation process was used to produce an organic fertilizer, OF, from
F. japonica using an innovative method of lactic fermentation by addition of EM Bokashi fermentation inoculum, complementary to some mechanical [
11,
16], biological [
14], and combined approaches [
41] already used to manage this invasive species.
Compared to classical aerobic thermophilic composting, fermentation has several advantages. The lack of aeration and mixing during the process; low energy input; further, no heat loss is produced, and no or few harmful gaseous compounds (e.g., CO
2, NH
3, and N
2O) are released into the air, and therefore more energy and plant nutrients remain in the fermented organic matter [
26]. The high level of acidity in OF, together with the inclusion of a probiotic lactic‒microorganisms consortia, was important to protect the material against further microbial degradation [
42,
43]. Furthermore, there is no loss of water-soluble K, as the fermentation process is conducted in a closed system where leaching does not occur [
26]. Besides organic acids, secondary products are formed, such as enzymes, amino acids, and vitamins, which have a stimulating effect on both soil microorganisms and the cultivated crops [
44]. Due to its production method, the C footprint of the fermented OF is lower than that of similar fertilizers (such as manure and compost), making OF an interesting alternative plant nutrient for reducing CO
2 emissions from agriculture [
26,
45].
Pathogenic microorganisms (e.g.,
Escherichia coli and bacteria of the genus
Salmonella) in fermented OF, moreover, do not reproduce due to acid environment (pH of OF was 4.6), making OF a safe product from a human-health perspective [
26]. The disadvantage of this treatment is that during fermentation, no high temperature is generated [
26], which would destroy the seeds of
F. japonica or inhibit their germination. Therefore, the plant material harvested for the production of OF should not include viable seeds or propagules with high reproductive capacity (with harvesting done in the flowering phase at the latest).
The experimental results prove the hypothesis that the OF produced from F. japonica could indeed successfully replace farmyard manure, which is not easily obtainable in urban environments.
The yields of Chinese cabbage mass of the whole heads obtained in fertilization experiments in some other studies are partially comparable to ours. Staugaitis et al. [
46] achieved a similar yield of 50 t ha
−1 with treatment N0 and 76 t ha
−1 with treatment N225, although the Chinese cabbage variety and the pedo-climatic factors were not the same as those in our study. The same statistically significant differences in the masses of marketable heads were observed between treatment groups, as reported by Staugaitis et al. [
46]. Dry matter content of Chinese cabbage plants was high, 9–11%, but comparable to the figures reported by Kosson et al. [
47]. The variability observed reflects a combination of abiotic factors (climate) and the soil system [
48,
49,
50,
51]. Based on previous research it would be interesting to focus further research on changes in the soil that might occur when using OF for a longer period (several years). It would be interesting to explore how the nutritive value of OF changes with an altered collection of raw
F. Japonica material for its preparation (different time of year) [
48,
49,
50,
52].
The fermented OF produced by
F. japonica contained 2.5-times more ash in dry matter than raw substrate from
F. japonica (
Table 2) and showed a well-balanced composition of main plant nutrients due to its high N sequestering and remobilization ability [
1], as well as due to lactic fermentation treatment, where no/little N losses occurred. The dry matter content in the OF was comparable to that of many composts and much higher than that in digestates [
53]. The quantity of N in fresh lactic-fermented OF was relatively high compared to average cattle farmyard manure but on a comparable level to solid poultry, and rabbit manure [
54,
55]. In comparison to compost and digestates, the relative concentration of N in OF was found to be much higher [
56].
The ratio of C/N in OF was relatively narrow (14.5:1), which indicates that N would be readily mineralized after incorporating the OF into the soil. According to Lazicki et al. [
57], organic matter with a C/N ratio >19:1 leads to N immobilization, whereas organic matter with C/N ratios <14:1 mineralizes N. This result is comparable to C/N ratios in well-composted plant material and higher than the ratios in many biogas digestates [
35].
Our results proved that after OF being applied and incorporated into the soil, the fermented organic matter was largely mineralized in the presence of oxygen and aerobic soil microorganisms, which resulted in high apparent N, P and K recoveries in Chinese cabbage yield (
Table 7). For an organic fertilizer of plant origin the contents of P and K in Chinese cabbage were relatively high. The ratio in kg t
−1 of fresh OF was in line with the needs of vegetable crops, which take up 4 to 8 times more K than P [
58]. However, the concentration of K (%) in the Chinese cabbage heads achieved in our research with OF (2.65% K) was a bit lower to the research of Pokluda [
59], where the average K was 3.1%, but this difference can be attributed to different pedo-climatic conditions.
Increasing yields due to increasing fertilization rates by OF accompany a certain dilution of N in plant biomass, which is due to the “law of diminishing yield increments”. Consistent with the findings of Mitscherlich [
60], the crop recovery efficiency decreases with an increase in fertilizer doses. The nutrient recovery efficiency is smallest for P and highest for K. Considering the large proportion of N initially organically bound in the OF, the recovery efficiency is shown here to be surprisingly high.
Studies reported that an average yield of Chinese cabbage (50 t ha
−1) [
47] takes up 26 kg ha
−1 P and 180 kg ha
−1 K. In our research, with 10 t OF ha
−1, we supplied 10 kg P and 100 kg K. Here, the crop recovery efficiency RE
N, P, K indicates the amount of (main) nutrient taken up by the crop and the remaining share in the soil. The crop directly took up quite a substantial fraction of the applied plant nutrients from OF (apparent recoveries for N159 were 55%, 48%, and 77% of N, P, and K, respectively), which is excellent for only two months of Chinese cabbage growth and to the amount expected for mineral N and K fertilizers. This result is even better than of mineral P-fertilizers, where the percentage recovery of fertilizer P calculated by the difference method normally ranges from 10 to 25% for a given crop in a given season [
61]. However, surpluses, particularly those of water-soluble nitrate, can be leached after harvest. If Chinese cabbage is the last crop during the year, a catch crop should be cultivated (e.g., winter cereals) to retain the nutrients available for the crops in the following year.
It can be assumed that OF in soil quickly mineralizes, releasing readily available N for the cabbage. In this study, the cabbage was not grown long enough to capture all N mineralized, as readily available mineral N was detected in the soil right after harvest (
Table 3). However, the soil mineral N after harvest was not elevated by OF application compared to the non-fertilized control, even under the highest N application, N317. Nitrate-N, which was the predominant mineral N formed in soil after the harvest, was high in content (32 to 34 mg NO
3-N/kg soil;
Table 3), which indicates high mineralization capacity of the experimental soil. Despite the high N-mineralization potential of the soil, the N input from OF was able to match the dynamic needs of the Chinese cabbage for N (good apparent N efficiency;
Table 8) and, consequently, achieved significantly higher yields without leaving excess mineral N in the soil after harvest compared to the Control, N0. This observation can be attributed to characteristic of OF: rather low C/N ratio, 14.5, and low content of fibers (lignocellulose), 8% (
Table 3), of which all being favorable for net N mineralization.
Table 9 shows the modeled required application of
F. japonica OF (1.59% N, 0.19% P, 1.00% K). We assumed garden soils with initial soil P and K supply in class C (optimal) and no plant-available mineral N readily available to meet the plant nutrient requirements of selected crops [
38] at a given expected average crop production yield based on the integrated plant nutrient management guidelines [
39]. Generally, N in OF was found to be the limiting factor among the main nutrients for plant growth. Moreover, the N availability for plants was significantly higher than that in comparable products [
62]. Traditional organic fertilizers used in organic farming often contain a P concentration above plant needs, which leads to an accumulation of P in soils, particularly if fertilization is quantified according to the required plant needs for N [
62].
The crop recovery efficiency levels of N in our short-term experiment were 55% (N159) and 37% (N317), and the average N159-N317 was 46% (
Table 8). OF, therefore, is a good source of N (under the condition that the organically bound N mineralizes as rapid as in the experiment), a moderate source of K, and a low source of P. In order to achieve an average yield of 5 kg/m
2 for the selected crops and to cover 100% of the N requirement of the crops, an application of 2 kg OF/m
2 would be necessary (
Figure 4).
This model application would cover 66.5% of P and 88.4% K average crop requirements (
Table 9). For some of the more N-demanding crops, such as cabbage, pumpkins, and tomato, higher OF application of 2–3.7 kg/m
2 is needed to achieve the targeted yield. Full coverage of N requirements would result in P and K oversupply for tomato and K oversupply for radicchio and paprika but would undersupply crop P requirement for onion, green beans, and carrot (P supply < 60%) and undersupply the K crop requirements of green beans and carrot (K supply < 60%) (
Table 9). Over the long run, some mineral fertilization could be applied to cover the gaps in the plant needs of P and K if mineral fertilization is permitted. Organic farming does not allow mineral N-fertilization but allows certain P- and K-fertilizers (which are not so strongly processed). In cases where P and K are oversupplied, vegetables with a longer growing season, as well as some of the plants grown as catch crops, possess the ability to access P and K resources from deeper soil layers [
63].
OF fertilizer treatments (N159, N318) showed sufficient nutrient uptake to meet the needs of fast growing vegetables. However, Staugaitis et al. [
46] warn that N applications greater than 180 kg ha
−1 can negatively influence the inner quality of Chinese cabbage in terms of its content of vitamin C, nitrates, and soluble solids. This negative aspect of high N fertilization is more probable when soluble N fertilizers are used. In a case of OF, N is organically bound and is released gradually. Even with the application of 318 kg N/ha, the inner quality of our cabbage was not degraded.
In organic vegetable production, the nutrient recovery achieved with frequently used organic fertilizers is lower; the mineralization rate in the first year of application for organically bound N in solid manure is 10–30%, for poultry manure up to 45% [
62]. Additionally, the nutrient dynamics after fertilization with OF should be further researched to explore nutrient dynamics under the repeated incorporation of OF in soil over several growing seasons and for different crops and soil conditions.
The approach adopted in this work was shown to be a viable alternative to manage the uncontrolled spread of
F. japonica in urban areas. The OF production process is simple enough to be repeated by small urban farmers and has the potential to be reproduced on a larger scale, by waste management companies. In this study, plant materials for the production of OF were cut and collected by hand. A more mechanized approach could be applied for the larger condensed
F. japonica patches that typically occur along the banks of highways, especially when the plants are mowed at least twice per year [
64]. This would ensure a high green-mass yield and an interruption in the reproductive cycle of the plant (pollen drift and seed formation) at the stage when the plant N content is expected to be highest, although precaution is necessary as some metallic trace elements, such as Zn, can transport to aboveground parts of
F. japonica [
65]. The approach taken in this study seems to be able to complement frequent (weekly) management of
F. japonica stands in parks, nurseries and riverbanks, where frequent cutting [
11] continues to be applied as a means of suppressing
F. japonica, although the long-term success of frequent mowing of
F. japonica as a management method remains to be proven [
66].