Effects of various nitrogen regimes on the ability of rapeseed (Brassica napus L.) to suppress littleseed canarygrass (Phalaris minor Retz.)

Background: Littleseed canarygrass (Phalaris minor Retz.) is one of the most troublesome invasive weeds infesting winter crops in Yunnan Province, China. Our previous study found that rapeseed (Bassica napus L.) was a logical candidate crop to control littleseed canarygrass in agroecosystems. Nitrogen (N) could impact plant community composition by altering competitive interactions, however, the effects of different N regimes on weed control ecacy of rapeseed were unknown. Here, we report the effects of different N regimes on the competitiveness of rapeseed with littleseed canarygrass and accompanying differences in photosynthetic characteristics. Results: The results showed that the rapeseed yield and its control ecacy on littleseed canarygrass were signicantly affected (P<0.05) under different N regimes, and the control ecacy of littleseed canarygrass by rapeseed increased rst and then decreased with the increase of basal nitrogen rates, while increasing topdressing N rates increased control ecacy of littleseed canarygrass by rapeseed. In fact, yield and weed control ecacy of rapeseed was most ideal when both basal and top-dressing N was 90 kg·ha -1 . We also found that N signicantly impacted the competitive ability of rapeseed to littleseed canarygrass and rapeseed had the highest competitive ability when both basal and top-dressing N was 90 kg·ha -1 . With the increase of basal nitrogen rates, competitive balance index (CB) of rapeseed increased initially but decreased beyond an optimal level. CB continually increased with increasing of topdressing N rates. Our research also showed level and period of N application had a signicant effect (P<0.01) on the photosynthetic rate (Pn) and chlorophyll content (Chl) of both rapeseed and littleseed canarygrass. Under the same N application regime, the Pn and Chl of littleseed canarygrass were higher than that of rapeseed in December, while the Pn and Chl of rapeseed was higher than that of littleseed canarygrass in February. Our study indicated that photosynthetic characteristics of rapeseed and littleseed canarygrass in different growth Our results demonstrated that optimal application of fertilizer N allowed rapeseed to produce strong yields and provide high ecacy in the control of littleseed canarygrass. We found that interspecic competition and photosynthetic characteristics of rapeseed and littleseed canarygrass could be greatly affected by different nutrient regimes. Littleseed canarygrass growth was more sensitive to basal N application than topdressing applications. Rapeseed had the strongest competitive ability for intermediate basal and topdressing N rates of 90kg·ha − 1 . Our study also indicated that photosynthetic characteristics of rapeseed and littleseed canarygrass at different growth stages varied in their sensitivity to varying N regimes, which showed that optimally planned nutrient regimes may provide a strategic tool for the ecological control of littleseed canarygrass. Finally, we recommend that future modeling or experimental studies on utilizing crops to control invasive weeds in agroecosystems should simultaneously consider impacts of N application on both crops and weeds to calculate optimum fertilizer rates to maximize crop competitiveness.

while increasing topdressing N rates increased control e cacy of littleseed canarygrass by rapeseed. In fact, yield and weed control e cacy of rapeseed was most ideal when both basal and top-dressing N was 90 kg·ha -1 . We also found that N signi cantly impacted the competitive ability of rapeseed to littleseed canarygrass and rapeseed had the highest competitive ability when both basal and top-dressing N was 90 kg·ha -1 . With the increase of basal nitrogen rates, competitive balance index (CB) of rapeseed increased initially but decreased beyond an optimal level. CB continually increased with increasing of topdressing N rates. Our research also showed level and period of N application had a signi cant effect (P<0.01) on the photosynthetic rate (Pn) and chlorophyll content (Chl) of both rapeseed and littleseed canarygrass. Under the same N application regime, the Pn and Chl of littleseed canarygrass were higher than that of rapeseed in December, while the Pn and Chl of rapeseed was higher than that of littleseed canarygrass in February. Our study indicated that photosynthetic characteristics of rapeseed and littleseed canarygrass in different growth stages differ in their sensitivity to N regimes, creating a dynamic competitive relationship.
Conclusions: Together, our results demonstrated that optimal application of fertilizer N could help rapeseed produce higher yields and greater weed control e cacy, suggesting that future modeling or experimental studies on utilizing crops to control invasive weeds should carefully consider both timing and placement of N.

Background
Invasive weeds in agro-ecosystems reduce the yield of crops, seriously affecting the export trade of agricultural products, and even threatening ecosystem safety and human health [1,2]. In order to meet these challenges, and to provide alternatives to control by herbicides or mechanical methods environmentally friendly approaches, which have been explored extensively over the past several decades. Ecological control using high value species (e.g., crop species grown locally, native species and/or cash crops) has emerged as a viable option for management of invasive alien plant species [3][4][5].
Compared with mechanical or chemical control methods, ecological control has generally been considered safer, economical, eco-friendly, and sustainable [6]. However, due to increasing demand for food, nitrogen (N) application to croplands has been on the increase [7], which may affect ecological control e cacy. Some studies have shown that increasing N application rates affect growth and control e cacy of native species on invasive weeds [8][9][10]. For example, increased N application was found to increase the invasiveness of the plant, Ambrosia artemisiifolia, as well as its ability to be suppressed by competition with native species [8]. Response to N levels varies widely by species, which explains although increased N in uences A. artemisiifolia favorably, other plants have yet stronger responses to N. Furthermore, increases in biomass due to increased N can make other nutrients less accessible. Given N is an indispensable nutrient for agricultural production, studies on the effects of various N regimes on the ecological control e cacy of native plant for managing invasive weeds are critical for future management of invasive weeds in agroecosystems.
Littleseed canarygrass (Phalaris minor Retz.), an annual weed native to the Mediterranean region [11,12], is one of the most harmful weeds of winter crops in the world. This weed has spread across Eurasia, South and North America, East and South of Africa and Australia [12,13]. This weed came to China from Mexico via the introduction of wheat in the 1970s and is presently among the most destructive invasive plants in temperate eld cropping systems of Yunnan Province, Southwest China [14][15][16]. Numerous methods have been developed to manage this weed, but the most effective control is generally achieved using herbicides [17][18][19]. Environmental issues stemming from the use of herbicides, along with the frequent occurrence of herbicide resistance, provide a good rationale for the implementation of alternative control measures [20]. As a promising alternative to traditional control methods, the use of rapeseed to control the invasive weed P. minor has recently been reported in China [21].
Rapeseed ranks second among almost all oilseed crops worldwide and is the rst-ranked oilseed crop in Yunnan Province [22]. Rapeseed competes strongly with weeds and also exhibits allelopathic activity, facilitating ecological control of weeds in farmland by rapeseed [16,21]. For example, one study found that Alopecurus japonicus Steud., Myosolon aquaticum (L.) Mocnch. and Lapsana apogonoides Maxim were signi cantly inhibited by rapeseed in farmland [23]. Our previous eld surveys found that population densities of littleseed canarygrass in rapeseed elds were signi cantly lower than those in other elds [17]. Subsequent studies examined the effects of rapeseed on littleseed canarygrass growth and reproduction [16]. In mixed culture, rapeseed exhibited greater competitive ability than littleseed canarygrass, and the total biomass, branch, panicles numbers and seed numbers of littleseed canarygrass were suppressed signi cantly by rapeseed [16,21]. As one of the key factors affecting crops, weeds and their interactions, N fertilizer has been widely applied in farmland to increase rapeseed yield.
However, the effects of various N regimes on the e cacy of rapeseed in controlling littleseed canarygrass are not known.
Building on our previous studies [15,16,17,21], the current research examined the in uence of various N application rates on control e cacy of rapeseed and littleseed canarygrass in a eld experiment in Yunnan Province, China. Our overall goals were to elucidate mechanisms by which rapeseed competes with littleseed canarygrass under different N regimes and explore more sustainable management Page 4/17 methods for littleseed canarygrass in agro-ecosystems. Speci cally, our objectives were to: (1) Evaluate impacts of different N regimes on rapeseed yield and its weed control e cacy; and (2) Measure competitive effects and photosynthetic characteristics of rapeseed and littleseed canarygrass under different N regimes.

Results
The impacts of littleseed canarygrass on the growth and yield of rapeseed under various nitrogen regimes The results showed that the aboveground biomass and yield of rapeseed in all mixed culture treatments were signi cantly lower (P < 0.05) than that in all monoculture treatments under the same N application rate ( Table 1). Effects of littleseed canarygrass on the growth and yield of rapeseed were signi cantly different (P < 0.05) under various N regimes. In monoculture, the aboveground biomass and yield of rapeseed increased with increasing N application rates. In mixed culture, the biomass and yield of rapeseed were highest for moderate basal N application rates, but the biomass and yield of rapeseed was highest for the highest topdressing N application rate. Under different N regimes, littleseed canarygrass could reduce rapeseed yield by 22.95% -54.60%, and yield loss was greatest (over 50%) with basal application rates of BN3. .

The inhibitory effects of rapeseed on littleseed canarygrass under various nitrogen regimes
The aboveground biomass and seed number of littleseed canarygrass were signi cantly inhibited (P < 0.05) by rapeseed under various N regimes (Table 2). In monoculture, the aboveground biomass and seed number of littleseed canarygrass increased with increasing N application rates. In mixed culture, the seed number was signi cantly decreased by the presence of rapeseed but seed production increased with increasing basal N application rates. Seed production in mixed culture by littleseed canarygrass decreased with increasing topdressing N application rates, however. Aboveground biomass of littleseed canarygrass did not differ between BN 1 and BN 2 treatments, but under the basal N application rate BN 3 , aboveground biomasses of littleseed canarygrass were signi cantly higher than for lower basal N rates ( Table 2). Rapeseed exhibited high control e cacy on littleseed canarygrass at low basal N rates but decreased with increasing basal N application rates. In mixed culture, seed number of littleseed canary grass was lowest at intermediate basal N application rates, whereas at highest basal N application rates control e cacy declined as little seed canary grass was the highest. Furthermore, within the highest basal N rates, seed production of the weed increased with increasing top dressing rates.

Competitive Interactions of rapeseed and littleseed canarygrass
The RY of rapeseed and littleseed canarygrass was signi cantly less (P < 0.05) than 1.0 in mixed culture, showing that the intraspeci c competition between the two plants was less than their interspeci c competition under different nutrient levels ( Table 3). The CB (competitive balance) index of rapeseed was signi cantly greater than zero in mixed culture, except at basal N level of BN 3 , which indicates that rapeseed was more competitive than littleseed canarygrass at low and moderate basal N levels, but sometimes less competitive at high basal N levels.
Under the same topdressing condition, with increasing basal N application rates, the RY and CB index of rapeseed rst increased and then decreased, and the RY of littleseed canarygrass rst decreased and then increased. However, with increasing topdressing N application rate, the RY and CB index of rapeseed increased, the RY of littleseed canarygrass declined.

Photosynthetic characteristics
In December, net photosynthesis rate (Pn) and chlorophyll content (Chl) of littleseed canarygrass were signi cantly higher than those of rapeseed (P < 0.05) in either monoculture or mixed culture (Figs. 1 and  2). With the increasing N application rates, the Pn and Chl of littleseed canarygrass increased; but Pn and Chl of rapeseed rst increased and then decreased in the mixed culture treatment. Although the Pn and Chl of rapeseed or littleseed canarygrass in mixed culture were lower than these parameters in monocluture, there were generally no signi cant differences for either species. However, for BN 3 treatments, the mixed culture rapeseed exhibited signi cant lower Pn and Chl levels by comparison to the other treatments( Figs. 1 and 2).
In February, the Pn and Chl of rapeseed were signi cantly higher than that of littleseed canarygrass (P < 0.05) in both monoculture or mixed culture (except the basal N application rate of BN3) (Figs. 1 and 2). In monoculture treatments, the Pn and Chl for both rapeseed and littleseed canarygrass gradually increased with increasing N application rates. In mixed culture treatments, Pn and Chl of rapeseed were highest at moderate basal N application rates, whereas Pn and Chl of littleseed canarygrass gradually increased as rates increased. However, Pn and Chl for both rapeseed and littleseed canarygrass in mixed culture gradually increased with increasing topdressing N application rates ( Figs. 1 and 2).

Discussion
Utilizing crops to control invasive weeds are regarded as an environmentally friendly approach to the management of invasive weeds in agro-ecosystems [6,25]. At the same time, due to the high demand for food, it is important to develop weed control methods to sustain crop yield. N is essential for plant growth and development and is widely used to improve crop vigor and productivity in agro-ecosystems [26]. Our previous studies found that rapeseed may be used to control littleseed canarygrass in agro-ecosystems [17]. It was generally hypothesized that increasing N rates could improve rapeseed yield and enhance the economic bene ts [24]. Furthermore, we theorized that different types of N application, such as basal or top-dressing applications might in uence responses of the crop and the weed as tested here. Crop and weed responses to N differ and are affected by competition, in uencing the outcome of interactions between rapeseed and littleseed canarygrass [16]. We found that increased topdressing N rates decreased rapeseed yield loss in mixed culture, but the yield losses in rapeseed were even more in uenced by basal N levels. For basal N application rates of BN 3 (150 kg·ha − 1 ), the yield loss of rape was more than 50%. However, the yield loss of rapeseed was less than 30% when for basal fertilizer rate BN 2 (90 kg·ha − 1 ). Our results suggested that weed control e ciency of rapeseed was greatly affected by N application levels, with basal N levels playing a leading role. Although at higher basal rates rapeseed yield was seriously reduced, intermediate N application rates may reduce the yield loss of rapeseed and could be successfully manipulated in favor of rapeseed, Competitiveness of crops plays an important role in determining the likelihood of success in the control of invasive weeds [4,5,21]. Although increased N levels can improve the competitiveness and plant growth of both crops and invasives [8,9], the relative effects vary depending on the particular species and other conditions, and adaptation for N uptake and utilization are related to the evolutionary history of invasive species and the arti cial selection of crops. .Our evaluation of competitiveness via seed production of littleseed canarygrass provided another view of the effectiveness of various treatments. Just as seen in the biomass measurements, rapeseed showed the highest control e cacy at intermediate basal N rates, while at the highest N rates littleseed canarygrass produced substantially higher seed numbers, resulting in a control e cacies in the 40% range, as compared to 67-80% at intermediate basal N rates, depending on accompanying top-dressing rates, with increased topdressing N rates improving the competitive ability of rapeseed. Thus the N level and type of fertilization must be regulated carefully to favor rapeseed over littleseed canarygrass.
As an essential nutrient element for plant growth, N is integral for plant photosynthesis [27]. Higher rates of photosynthesis can lead to increased growth rates, biomass accumulation and overall production [28].
Our current study showed that net photosynthesis rate (Pn) and chlorophyll content (Chl) of littleseed canarygrass were signi cantly higher than those of rapeseed (P < 0.05) in December. At that point in the season, the Pn and Chl of littleseed canarygrass increased with increased N whereas by comparison Pn and Chl of rapeseed did not respond as well to the higher fertilizer levels. However, in February, the Pn and Chl of rapeseed were signi cantly higher than those of littleseed canarygrass (P < 0.05). At this point, increasing basal N continued to produce higher littleseed canarygrass Pn rates, but only increased rapeseed rates over the transition from BN1 to BN2; however, topdressed N consistently increased Pn and Chl in both species. Thus, in the seedling stage, N demand of littleseed canarygrass was stronger than that of rapeseed, with the result that higher N levels did not increase rapeseed productivity as much as the invasive plant littleseed canarygrass. However, by the reproductive stage, increasing topdressing N improved did improve the productivity of rapeseed.
Differences in nutrient requirements of plants at different growth stages is a common phenomenon [29]. If carefully matched to N demands at various plant growth stages, fertilizer availability could also be manipulated to provide ecological management of invasive alien plants. In our study we found that the net photosynthesis rate (Pn) of rape at seedling stage is weaker than that of littleseed canarygrass, while it is stronger at propagation stage. Our nding of a relatively low net photosynthesis rate (Pn) of rapeseed at the seedling stage, the e cacy of rapeseed in controlling littleseed canarygrass could be improved by timing the fertilizer application appropriately. By using fertilization strategically, e.g., by utilizing topdressing methods, N could be applied to deliberately to improve the competitive ability of rapeseed over invasive plants like littleseed canarygrass. Control indices are useful to developing clear measures for evaluating invasive plant control e cacy [30]. Our work was the rst study to examine rapeseed yield and littleseed canarygrass weed seed number to evaluate the effect of N regime on their competitive relationship. Our measure of invasive plant control e cacy involved pitting yield of the crop (rapeseed) against the seed production of the weed (littleseed canarygrass), where yield is the measure of crop success and weed seed is essential for the establishment and spread of weed populations. Also key to our approach was showing how the two plants competed at different densities using a De Wit replacement series experiments [25] using biomass to develop an index to evaluate plant interspeci c competition [31] as well as yield measures. The data from the De Wit replacement series enabled us to calculate competitive balance, which revealed that interspeci c competition had a large effect on both species, larger than the effects of intraspeci c competition in both cases. It was also clear that topdressing N application favors rapeseed over littleseed canaraygrass, whereas basal N applications only favor rapeseed to an intermediate level, but at higher levels the competitive balance is tipped in favor of littleseed canarygrass.
With invasive weeds continuing to threaten food safety and agro-ecosystem sustainability, ecological control using crops or native species may provide safe, economical, and environmentally sustainable solutions for invasive weed management [24,31]. Therefore, choosing crop species with strong competitiveness, high economic value and suitable for large-scale planting have formed the core issues of this study [5]. However, the interspeci c competition between plants is dynamic. Environmental factors such as N availability may affect the interspeci c relationship between alternative crops and invasive weeds as demonstrated in our study.

Conclusion
Our results demonstrated that optimal application of fertilizer N allowed rapeseed to produce strong yields and provide high e cacy in the control of littleseed canarygrass. We found that interspeci c competition and photosynthetic characteristics of rapeseed and littleseed canarygrass could be greatly affected by different nutrient regimes. Littleseed canarygrass growth was more sensitive to basal N application than topdressing applications. Rapeseed had the strongest competitive ability for intermediate basal and topdressing N rates of 90kg·ha − 1 . Our study also indicated that photosynthetic characteristics of rapeseed and littleseed canarygrass at different growth stages varied in their sensitivity to varying N regimes, which showed that optimally planned nutrient regimes may provide a strategic tool for the ecological control of littleseed canarygrass. Finally, we recommend that future modeling or experimental studies on utilizing crops to control invasive weeds in agroecosystems should simultaneously consider impacts of N application on both crops and weeds to calculate optimum fertilizer rates to maximize crop competitiveness.

Study Species
Littleseed canarygrass (Phalaris minor Retz.) is widely distributed in subtropical and temperate regions in Yunnan Province, Southwest China [14,17]. Since 2013, littleseed canarygrass seeds have been collected from wheat elds in Songming County of Yunnan Province and propagated in the glasshouse of the Agricultural Environment and Resource Research Institute, Yunnan Academy of Agricultural Sciences, China. The average weight of 1000 seeds was 1.49 ± 0.05 g and the germination rate using culture dish lter paper method was 91.8%, as tested before the experimentation.
Rapeseed (Brassica napus L.) is a dominant oil crop in subtropical and temperate regions of Yunnan Province [22]. Our previous studies showed that rapeseed variety Yunyou No. 2 had strong competitive ability and could be used to control littleseed canarygrass in agroecosystems [24]. In this experiment, the seeds of Yunyou  [17]. This area is characterized by a subtropical and temperate monsoon climate. Rainfall averages 1000-1300 mm per year and the annual mean temperature is 14.1°C [17]. The soil type of test site is yellow-brown, with a total N content of 1.32g·kg − 1 , Olsen P content of 29.5 mg·kg − 1 , and available K content of 86.4 mg·kg − 1 , respectively.
The experiment was a split-plot design with basal N fertilization rates as the whole-plot factor (four replications for each N application rate treatment), topdressing N fertilization rates and plant ratios were split-plot factors, utilizing a de Wit replacement series method [32].  four replicates per ratio and per nutrient level (total n = 4 replicates × 3 ratios × 9 nutrient levels = 108). A 1.0 m border was constructed between plots and each plot was fenced with 0.5 m high glass panels to avoid being disturbed by herbivores The net photosynthetic rate (Pn) measurements on leaves for rapeseed and littleseed canarygrass conducted using a Portable Photosynthesis System (LI-COR Biosciences LI-6400XT, Lincoln, Nebraska, USA), between 10:00 am and 16:30 pm, with a 6400-02 LED source and 1000 µmol m − 2 s − 1 photosynthetically active radiation [33]. During sampling, CO 2 concentration in surrounding air, air temperature and relative humidity (RH) in the chamber were measured. In December 2018, two months after transplanting, 40 fully expanded leaves ( ag leaf and the top second leaf) of each species were randomly sampled from each plot and immediately scanned using an LI-6400XT [34]. Then, the leaves were cleaned for chlorophyll content determination. We cut 0.1g pieces from fresh leaves avoiding the main vein, and soaked the leaf fragments in 25 ml 95% ethanol at room temperature for 48 h. The detector was set at 665 nm in order to calculate the total chlorophyll content (Chl) [28]. In February 2019, four months after transplanting, twenty plants of each species were selected randomly and forty fully expanded sun leaves ( ag leaf and the top second leaf) of each species were sampled for net photosynthetic rate and chlorophyll content, following the same method as above.
Plants were manually uprooted and then cut at ground level for determination of aboveground biomass. Fresh plants were heated for 30 min at 105°C to halt metabolic processes, and then dried at 80°C in a forced-draft oven until reaching a constant weight before weighing. Rapeseed yield and littleseed canarygrass seed number were determined in a 2m 2 area in each plot; rapeseed yield was adjusted to a moisture content of 10.0%.

Data analyses
To evaluate the effect of littleseed canarygrass on rapeseed yield under various N regimes, the yield losses of rapeseed were determined from rapeseed yield comparing monoculture versus mixed culture according to the formula: the yield reduction rate of rapeseed(%)=(1 yield in mixed culture / yield in monoculture)×100%. To evaluate the ability of rapeseed to compete effectively with littleseed canarygrass under various N regimes, we determined the e cacy of littleseed canarygrass suppression by comparing its seed numbers in monoculture versus mixed culture according to the formula: control e cacy of littleseed canarygrass (%)=(1 seed number in mixed culture / seed number in monoculture)×100%.
Relative yield (RY) per plant [35] and competitive balance index (CB) [36] were calculated from the nal aboveground biomass obtained for each species in each plot. Relative yield per plant of species a or b in a mixed culture with species b or a was calculated as RY a = Y ab /Y a or RY b = Y ba /Y b . Competitive balance index was calculated as CB a = In (RY a /RY b ), where Y ab is the yield for species a growing with species b (g/individual), Y ba is the yield for species b growing with species a, Y a is the yield for species a growing in pure culture (g/individual), and Y b is the yield for species b growing in pure culture. Values of RY ab measure the average performance of individuals in mixed cultures compared to that of individuals in pure cultures. An RY ab of 1.00 indicates species a and b are both equal in terms of intraspeci c competition and interspeci c competition. An RY ab greater than 1.00 means intraspeci c competition for species a and b is higher than interspeci c competition, and an RY ab of less than 1.00 implies intraspeci c competition of species a and b is less than interspeci c competition [37]. Values of CB a greater than 0 indicate that species a is more competitive than species b [38].
The rapeseed yield and its yield reduction rate, seed number of littleseed canarygrass and its control e cacy, aboveground biomass, physiological (Pn) and chlorophyll content of rapeseed and this weed were analyzed by analysis of variance (one-way ANOVA) using IBM SPSS 23.0 software (Armonk, New York, USA). The F and partial eta squared statistics were calculated considering density ratio and N level with their interaction as factors at a 5% level of signi cance. Relative yield from each mixed culture were compared to the value of 1.00 using t-tests (α = 0.05), and values of CB for deviation from 0 using a paired t-test.