Introduction

The practice of rainfed wheat crop under the system of permanent bed planting has been increasing. The success of this technology has been based on the proper application of three basic principles; minimal movement of soil, crop rotation and permanent soil cover crop residues (^{Hobbs et al., 2010}; ^{Sommer et al., 2014}). The benefits from the application of these principles generally relate to performance improvement (^{Govaerts et al., 2005}) and physical, chemical and biological soil attributes (^{Limón-Ortega et al., 2002}).

However, previous research has shown that conservation tillage systems can create conditions that increase the incidence of fungal species in the soil (^{Steinkellner and Langer, 2004}). The reason is that the crop residues are important source of inoculum for ailments such as Fusarium spp. (^{Dill-Macky, 2008}; ^{Wegulo et al., 2015}). For example, depending on the crop rotation, ^{Dill-Macky and Jones (2000)} found that the incidence and severity of Fusarium spp. Is higher wheat in rotation with corn in rotation with soybeans. Similarly, it has been shown that the continuous rotation of wheat-corn in combination with favorable climatic conditions can result in the accumulation of inoculum (^{Landschoot et al., 2013}). In contrast, other research reports have concluded that the role of tillage on diseases is not clear recognition that a healthy soil with high microbial diversity plays an antagonistic role to soil-borne pathogens (^{Hobbs et al., 2010}).

On the other hand, most yield losses in both wheat and barley is caused by species of Fusarium fungi, which species Gibberella zeae (Fusarium graminearum), culmorum and Gibeberela avenacea (Fusarium avenaceum) are the most important (^{Landschoot et al., 2013}; ^{Wegulo et al., 2015}). Also, these species can survive saprophyte way for several years in the soil, particularly in crop residues (^{Leplat et al., 2013}). In contrast, the system of permanent bed planting beds with conventional tillage or show a marked reduction in some foliar diseases as a possible consequence of micro climatic conditions resulting from the change in the orientation of plants (^{Sayre and Hobbs, 2004}).

Other factors may also handling promote or inhibit the development of specific communities Fusarium spp. is weed control herbicide glyphosate and source of N fertilizer (^{Leplat et al., 2013}). Moreover, some studies have reported that the assessment of NDVI from reflectance measurements at different stages of development using the GreenSeeker sensor is related to the concentration of N and final performance (Teal et al., 2006). As a result, the potential wheat yields have been predicted from the early stage of development in order to maximize the efficiency of applications based fertilizer N (Teal et al., 2006; Martin et al., 2007). However, no reports of a direct relationship between NDVI readings and root diseases.

Because of the devastating nature of Fusarium spp., is essential for producers to have strategies to mitigate yield losses by this pathogen. These strategies include a combination of cultural practices such as N management, planting resistant or tolerant varieties, chemical and biological control, use of forecasting systems and harvesting strategies. Therefore, the objective of this study was to assess the incidence of fungal populations in wheat roots under permanent bed planting and doses of N and its influence on wheat yield, yield components and absorption of N.

Materials and methods

The wheat plant roots were collected in the 2014 cycle in a field trial under permanent bed planting cycle established in 2008, after nine agricultural cycles in the experimental field "Valle de Mexico" (CEVAMEX) of INIFAP. The incidence of disease was estimated in samples collected four plants roots 15, 30, 60, 90 and 120 days after the emergency treatment and four N (0, 20, 40 and 60 kg ha^-1). Each of these treatments was nested within three doses of N (0, 60, 120 kg ha^-1) applied to corn in the previous cycle which resulted in a total of nine experimental units per dose of N applied to wheat. The incidence of each species was determined as the number of infected roots through sampling. To identify the species present, the roots were rinsed after sampling with distilled water, dipped in a solution of 5% NaOCl for 30 s and rinsed again for a time equal in distilled water. The samples were dried in sterile filter paper and placed in Petri dishes in a medium with PCNB agar 50 mg streptomycin sulfate and 18 g per liter of abamectin (Agrimec^TM) within growth chambers. After six days, the Petri dishes were examined when fungus species could be identified by direct microscopic observation according to the morphological description of ^{Zyllinsky (1984)}.

The field experiment consisted of planting wheat after corn on a biannual rotation under rainfed conditions under the system of permanent bed planting with residue of both crops and stubble on the surface. The granular urea (460-0) was used as a source of N. The N treatments were maintained in the same plot each year and applied in-band over the beds and incorporated partially during the seeding operation. For weed control before planting annually it applied herbicide glyphosate. Postemergence control of broadleaf and grass tillering was conducted during the application of 10 g ha^-1 of prosulfuron (sulfunilurea) and 300 g ha^-1 clodinafop (aryloxyphenoxypropionates), respectively.

The grain yield estimate was performed by harvesting the two central beds. The total N uptake was estimated at 50 stalks randomly collected in each treatment. These samples were dried and threshed to determine the percentage of N in grain and straw with NIR 6500's spectrophotometer (FOSS, Denmark). The reflectance of the plants was evaluated in three stages of development (boot stage, anthesis and grain filling) using the optical sensor GreenSeeker^TM (NTech Industries Inc.) to estimate the NDVI. Readings were taken on two central beds at a height of 0.9 m on plants and oriented perpendicular to the rows of plants.

The data were analyzed using the SAS program. The estimated incidence relations with measurements in plants and NDVI readings were analyzed using the PROC MIXED option and were considered significant when p< 0.05.

Results and discussion

The weather conditions for the agricultural cycle in which this study was conducted were a little cold and rainy than the historical average. The historical average maximum temperature in the study area assessed from July to September is 25.1 °C and while the average for the cycle 2010 was 24.2 °C. The amount of historical rainfall for this period is 331 mm while for 2010 cycle was 375 mm. Under these weather conditions the resulting average yield was 1 590 kg ha^-1.

The 14 species of fungi (Table 1) were identified; ten of these were pathogens patógenos (Fusarium spp. and Helminthosporium spp.) and four saprophytes (Alternaria, Rhizopus, Verticillium y Penicillium). Among the pathogenic species that had the highest incidence was found mainly Cochliobolus sativus (Helminthosporium sativum) and Fusarium moniliforme; however, the first species was the most frequently isolated. The incidence of these two species was followed Helminthosporium turcicum and Fusarium oxysporum and F. graminearum. These isolations were achieved after nine years of establishment of bed planting system when a significant decline was already expected in the incidence (^{Govaerts et al., 2006}).

The incidence of these five pathogenic species was significantly affected by the dose of N applied to wheat while the previous cycle maize had no effect on any of them (Table 2). However, the incidence of most of these fungi through N rates applied to wheat was inconsistent, except for Fusarium moniliforme and F. oxysporum. In the case of the first, the incidence increased with application of N from 0 to 40 kg ha^-1 and declined from 40 to 60 kg ha^-1 (Figure 1). Conversely, the incidence of Fusarium oxysporum was consistently reduced to the extent that the dose of N was increased from 0 to 60 kg ha^-1 which coincides with the report Warren and Kommendahl (1972). These results indicate that the N source is not the only factor affecting management community Fusarium spp. (^{Leplat et al., 2013}) but also the dose of N applied as urea. Moreover, the irregularity in the incidence through N rates for the rest of the species identified indicated that it depends on factors other than the amount of fertilizer applied.

Figure 1 Incidence of Fusarium moniliforme and F. oxysporum on wheat in rotation with maize in response to the application of four doses of N under permanent beds. 

In order to identify the effect on the performance of these two species, Fusarium moniliforme and F. oxysporum, a regression analysis of these variables for each dose of N. The results Fusarium moniliforme was conducted they indicated generally a significant difference for the intercept coefficient and slopes between doses of N (Table 3). The opposite was obtained for Fusarium oxysporum whose incidence was not related to any of the parameters evaluated plant. This lack of relationship, including the incidence of F. graminearum has been reported as devastating pathogen (^{Wegulo et al., 2015}) suggests that the N and cycle environmental conditions were not factors that promote the occurrence of these two species. Another suggestion is that the effect of the system is planted in reducing the incidence (^{Govaerts et al., 2006}) is specific only for some species.

According to regression analysis, the slope between the incidence of Fusarium moniliforme and performance indicated that the cup this reduction was greater for the dose of 60 kg N ha^-1 (Figure 2). It is also important to note that although the dose of 40 kg N ha^-1 resulted in a higher incidence of this pathogen, its effect on the yield was virtually null. Coincidentally, the last dose of N has been previously reported as adequate to obtain acceptable yields of rainfed wheat in Valles Altos de Mexico (^{Limón-Ortega and Sayre 2012}). Therefore, strategies to address some of the root diseases caused by Fusarium spp. should include cultural practices (^{Wegulo et al., 2015}) such as N management (^{Lori et al., 2009}).

Figure 2 Relationship between wheat yield and incidence of Fusarium moniliforme in four doses of N (0, 20, 40 and 60 kg ha^-1). 

The N handling, moreover, has been identified as the strategy that promotes the number of pins which is one of the most important components of this crop performance (^{Phillips et al., 2004}). In this study, the results indicated that the incidence of the pathogens identified had no substantial effect on the number of spikes, except Fusarium moniliforme (Table 3). In general, the higher rate of reduction in this component, according to the regression slope was observed at the highest dose of N (Figure 3) while for the dose of 40 kg ha^-1 rate was zero even though the incidence was higher for this dose. This result indicates the importance of dosing of N to reduce the incidence effect in plants.

Figure 3 Number of spikes per m² and its relationship with incidence of Fusarium moniliforme in four doses of N (0, 20, 40 and 60 kg ha^-1). 

Moreover, evaluation of the total amount of N absorbed a crop is used to estimate the efficiency of use of this element (^{Samonte et al., 2006}). This efficiency is estimated globally 33% (^{Raun and Johnson, 1999}) depending on the operational definition applied. In general, this efficiency is reduced by the use as the dose of N is increased as a result of the reduction in absorption of this element (^{Halvorson et al., 2004}). Regression analysis in this study revealed that the incidence of Fusarium moniliforme is negatively associated with the absorption of N according to the dose applied (Table 3). The absorption of N to the lowest incidence (intercept) and cup highest reduction (slope) was observed for the dose of 60 kg N ha^-1. In contrast, the dose of 40 kg N ha^-1 had an intermediate initial absorption but without effect by the increased incidence (Figure 4). According to these results, the application of urea as a source of N to a different crop needs doses, promotes the development of diseases in the root which prevents proper absorption of N affecting the efficiency of use (^{Lori et al., 2009}). According to Figure 4, the difference between 60 and 40 kg N ha^-1 to low incidences of Fusarium moniliforme, an increase in the absorption of 35 kg N ha^-1 about which is substantial to the use efficiency improves N.

Figure 4 Total absorption of N and its relationship with the incidence of Fusarium moniliforme in four doses of N (0, 20, 40 and 60 kg ha^-1). 

Other factors also affect communities of fungi are weather conditions (^{Lori et al., 2009}) which are combined with crop residues in conservation tillage (^{Wegulo et al., 2015}). According to the climatic conditions of the area of this study, the relatively high precipitation and low average temperatures, may have been the conditions that promoted the incidence greater or lesser degree, depending on the dose of N of Fusarium species identified (^{Steinkellner and Langer 2004}).

Moreover, the association between the incidence of Fusarium moniliforme and NDVI readings collected at three stages of development (boot stage, flowering and grain filling) indicated that it is feasible to predict the presence of this pathogen from the reflectance of the crop. This result was observed for treatment only 60 kg N ha^-1 (Figure 5) with significant differences only in the intercept coefficient (Table 4). In general, to the extent that reflectance readings were greater, the incidence of this pathogen was lower. However, although generally the NDVI readings have been used to monitor the status of N in plants and forecast performance (^{Raun et al., 2005}), the results of this study indicate that the reflectance evaluated as NDVI crop can also be used to be associated with the incidence of pathogens in the root when the dose of N is excessive.

Figure 5 Relationship between NDVI readings collected at three stages of development (boot stage, flowering and grain filling) with the incidence of Fusarium moniliforme to 60 kg N ha^-1. 

Conclusion

Pathogen species were identified most frequently correspond to Fusarium spp. and Helminthosporium spp. The incidence in most species are affected by the application of N, however, only a regular consistency was observed across the dose of N in the species of Fusarium moniliforme and F. oxysporum. The incidence of the first species is maximized at the dose of 40 kg N ha^-1 but this issue did not have substantial effects on performance, spikes per m² or absorption of N. According to previous studies, this dose of N has been considered the right to obtain acceptable yields of rainfed wheat in Valles Altos. The negative effects on the incidence of Fusarium moniliforme on these variables in plants were observed at the dose of 60 kg N ha^-1 indicating the importance of N management as a factor to reduce the effect of impact on performance. In contrast to this species, the incidence of Fusarium oxysporum was reduced to the extent that the N rate increased from 0 to 60 kg ha^-1 without presenting any substantial effect on performance, spikes per m² or total absorption of N.

Moreover, this study demonstrated that the incidence of Fusarium moniliforme to high doses of N reduces the total absorption of this element which can be a cause of the low efficiency of use of N. Moreover, the collection of readings NDVI at different stages of development and high doses of N proved to be a promising tool to detect pathological problems in the root system.