Biomass flow and water efficiency of cactus pear under different managements in the Brazilian Semiarid 1

Balanced fertilization is an important factor in optimizing the growth and productivity of crops. This study aimed to evaluate biomass flow and rain use efficiency in cactus pear cv. Gigante fertilized with combinations of nitrogen (10; 70; 100; 130 and 190 kg ha-1 year-1 N) and phosphorus (10; 70; 100; 130 and 190 kg ha-1 year-1 P2O5), in two harvest frequencies (annual and biennal), in the regions of Quixadá and Tejuçuoca, in Ceará. This was a split-plot randomized block design, with four replicates. In Quixadá and Tejuçuoca, the N and P combinations and harvest frequencies influenced the productions rates of different cladode orders and rain use efficiency in cactus pear. In Quixadá, in the annual harvest, the forage production rate (FPR) was 10.3 kg ha-1 day-1 in the N/P2O5 combination at 190.0/190.0 kg ha -1 year-1 and in the biennal harvest, the maximum FPR was 4.6 kg ha-1 day-1 in the N/P2O5 combination at 114.6/136.8 kg ha -1 year-1. In relation to the annual harvest in the municipality of Tejuçuoca, the maximum FPR was 30.0 kg ha -1 day-1 in the N/P2O5 combination at 137.5/190.0 kg ha-1 year-1 and in the biennal harvest, the maximum FPR observed was 38.0 kg ha -1 day-1 with the N/P2O5 combination of 190.0/33.4 kg ha -1 year-1. The biomass productions rate and rain use efficiency are optimized with the balance of combined levels of nitrogen and phosphorus.


INTRODUCTION
Forage production under dryland regime in the semi-arid regions has been shown to be a very important aspect in the livestock exploitation, since under such conditions the cattle rancher faces a serious problem, which is the qualitative-quantitative deficit of the forage produced, since the constant dry periods associated with the consecutive years of low rainfall greatly alter the production and quality of the forage crop.In this sense, cactus pear is an important forage resource for animal feed in semi-arid regions (CAVALCANTI et al., 2008), once it is a crop with high productive potential in these regions, in addition to high tolerance to arid and semi-arid conditions, with CAM photosynthesis and high efficiency in water use (OLIVEIRA et al., 2010;SILVA et al., 2014).
Balanced fertilization is an important factor in optimizing the growth and productivity of forage plants (LOPES et al., 2011a, b;POMPEU et al., 2010).The literature contains other studies on the effects of nitrogen and phosphate fertilization on the structural and productive characteristics of cactus pear (ALMEIDA; PEIXOTO; LEDO, 2012;CUNHA et al., 2012;DUBEUX JÚNIOR et al., 2006).Besides fertilization, other factors including the frequency and intensity of harvests (ALVES et al., 2007;FARIAS et al., 2000) and cultivation regions (DUBEUX JÚNIOR et al., 2006) influence biomass production of cactus pear.
Studies demonstrated the variation in the magnitude of cactus pear responses to fertilization and, in general, involved one or a few factors.This fact demonstrates the need for comprehensive research on the morphophysiological responses of cactus pear to multiple factors.
The focus of the present research will enable the mapping and quantification of the integrated responses of cactus pear components to environmental and management factors, encompassing different soil and climatic conditions (distinct regions), fertilization (nitrogen and phosphorus combinations) and harvesting frequencies (annual and biennal).
In the light of the above, this study was conducted with the purpose of evaluating the biomass flow characteristics and rain use efficiency in cactus pear cv.Gigante, fertilized with nine combinations of nitrogen and phosphorus under two harvest frequencies, in the semiarid regions of Quixadá and Tejuçuoca.

MATERIAL AND METHODS
The experiment was conducted in the municipalities of Quixadá and Tejuçuoca, State of Ceará, Brazil.In Quixadá, the experiment was conducted at Lavoura Seca Farm, belonging to the Agricultural Sciences Center, Universidade Federal do Ceará (UFC).In Tejuçuoca, it was conducted at Quilombo Farm, privately owned.The municipality of Quixadá has BSw'h' climate, hot semi-arid, and Tejuçuoca, Aw climate, tropical with dry season, according to Köppen classification (1948).
Daily rainfalls were recorded in both regions (Figure 1).The temperature (T) and relative humidity (RH) of the experimental period were also recorded.In Quixadá, mean values of 27.0 °C and 58.8% were recorded for T and RH, respectively, during the exprimental period.Data were obtained at the Agroclimatological Station of the Federal University of Ceará (Quixadá) and at the Agroclimatological Station of FUNCEME (Tejuçuoca).In both municipalities, the experiment began in 2011 and followed similarity regarding the evaluated treatments.
At the beginning of the experiment, the soil of both regions was analyzed for physical-chemical characterization at the depth 0.0 -20.0 cm (Tables 1  and 2).
After harvesting in the field, cladodes were allowed to rest in the shade for fifteen days (LOPES  Furrows were manually opened, using cutter mattock, narrow hoe and hoe, obeying the average depth of 30 cm and 2.0 m spacing.Cladodes were planted at the depth that covered 2/3 of them, at 2.0 m x 0.10 m spacing, recommended to meet the density of 50,000 plants ha -1 .
Fertilization was carried out during the rainy period.The annual dose of phosphorus, as single superphosphate, was applied at once, at the time of planting for the first year and when the rainy period began for the other years of cultivation.At this time, micronutrients were applied at a dose of 50 kg ha -1 FTE BR-12.Calcium and sulfur balancing was performed for all treatments, using agricultural gypsum and M. N. Lopes et al. calcitic limestone, based on the highest dose of single superphosphate.
The annual dose of nitrogen, as urea, was made available in three fractions, with a 20 day-interval between applications.In all applications, urea was diluted in water, applying 1.0 liter of solution for each 4 m length of the cactus pear row, totalizing 3.0 liters per plot.Dilution was carried out in order to better uniformity, due to the small volume of fertilizer per plot in some doses, making it difficult to apply in solid form.
After the establishment of the cactus pear plantation, crop treatments were carried out, whenever necessary, to maintain the crop in good growth conditions.
With respect to biomass flow, the following variables were determined: production rates of cladodes of the first (PRFOC, kg ha -1 day -1 ), second (PRSOC, kg ha -1 day -1 ), third (PRTOC, kg ha -1 day -1 ), fourth (PRFoOC, kg ha -1 day -1 ) and fifth order (PRFiOC, kg ha -1 day -1 ); determined from the division of the pre-dried biomass of cladodes of each order per hectare by the evaluation period; forage production rate (FPR, kg ha -1 day -1 ), determined from the division of total pre-dried biomass of all orders of cladodes per hectare by the evaluation period; harvested forage production rate (HFPR, kg ha -1 day -1 ), determined from the division of the pre-dried harvestable biomass of cladode orders above the cutting height per hectare by the evaluation period.The pre-drying of the cladodes was carried out at 55 °C to constant biomass.
Rain use efficiency (RUE) for total forage biomass pre-dried at 55 °C (TFB), in kg ha -1 mm -1 , was estimated from the division of TFB by the cumulative rainfall (mm) during the cycle.
Data were subjected to analysis of variance, test of comparison of means and multiple regression models.Qualitative factors were compared by the Scott-Knott means comparison test at 5% probability level.Quantitative factors were analyzed in multiple regression models (up to 10% probability level).The statistical genetic analysis system (SAEG 9.1, 2007) was used as a tool to aid analysis.

RESULTS AND DISCUSSION
When analyzing production rates and rain use efficiency in cactus pear, interactions were found between nitrogen and phosphorus combinations and harvest frequencies in Quixadá and Tejuçuoca (Table 3).Thus, the  Harvest frequency (HF); Production rates of first (PRFOC), second (PRSOC), third (PRTOC), fourth (PRFoOC) and fifth (PRFiOC) order cladodes; significant at 0.1% (***), 1% (**), 5% (*) and 10% ( ∆ ) probability combined effect of N and P on the studied variables was evaluated at the two harvest frequencies.The responses of the cactus pear to harvest frequencies were also analyzed.
The highest rates of biomass production of the different orders of cladodes in response to the N and P M. N. Lopes et al. supply for different harvesting frequencies, in Quixadá and Tejuçuoca, resulted from the positive effect of both on plant growth and development, accelerating the biomass increase by increasing the production of new cladodes and biomass gain per cladode as a function of the stimulus to cell division and elongation (VOLENEC; NELSON, 1984), contributing to the three-dimensional increase of each cladode order, with higher density of biomass per cladode, corroborating with the production rates of the whole plant.
In this context, given the importance of the nutrients mentioned for the greater gain of biomass by the forage crop within the same time of growth, it is also worth mentioning, with respect to the beneficial effects of nutrients, that both nitrogen and phosphorus are essential for the development of plants; phosphorus is essential for cell division, reproduction and plant metabolism, and nitrogen is part of all amino acids, which represent the components of proteins (BISSANI et al., 2008).In the present study, fertilization, mainly with nitrogen, is an important practice for optimization of the photosynthetic processes of CAM plants (NERD;NOBEL, 1995;NOBEL;BARRERA, 2002), making it a fundamental point for the maximum growth and production of forage biomass under different conditons of environment and management.
In Tejuçuoca, in the annual harvest, the maximum FPR of 30.0 kg ha -1 day -1 was found with the dose of 137.5 kg ha -1 year -1 N combined with the highest dose of P (190.0 kg ha -1 year -1 P 2 O 5 ).In the biennal harvest, the highest FPR (38.0 kg ha -1 day -1 ) was observed with the highest dose of N (190.0kg ha -1 year -1 ), combined with 33.4 kg ha -1 year -1 P 2 O 5 .In the annual harvest, the maximum HFPR at 20.0 kg ha -1 day -1 was verified with the dose of 133.6 kg ha -1 year -1 N combined with 142.7 kg ha -1 year -1 P 2 O 5 .In the biennal harvest, the maximum efficiency in HFPR (31.8 kg ha -1 day -1 ) was reached with the highest dose of N (190.0kg ha -1 year -1 ) combined with 65.1 kg ha -1 year -1 P 2 O 5 .
In Quixadá and Tejuçuoca, in annual and biennal harvests, the responses of forage production rate and harvestable forage production rate to the combined N and P supply is the final product resulting from the integration of the responses of the production rates of the different orders of cladodes of cactus pear, with the difference that the final FPR response to the combined N and P doses resulted from the performance of these nutrients on PRFOC, PRSOC and PRTOC in Quixadá, against the occurrence of PRFOC, PRSOC, PRTOC, PRFoOC and PRFiOC in Tejuçuoca.Regarding the HFPR in response to the N and P combinations, there was a combined effect of nutrients on the PRSOC and PRTOC in Quixadá, against their effect on PRSOC, PRTOC, PRFoOC and PRFiOC, in Tejuçuoca.
With regard to the pattern of FPR and HFPR responses in plants fertilized with combined doses of nitrogen and phosphorus, it is explicit that forage production, a reflection of the processes of plant growth and development, can be substantially improved with the use of fertilizers, especially by the supply of N and P, due to the positive effect of both on the biomass flow of cactus pear.In Quixadá, PRFOC and PRSOC were higher in the annual harvest in all combinations of N and P. The order of superiority in the PRTOC varied between annual and biennal crops, depending on the combination of N and P (Table 6).
Higher values of PRFOC and PRSOC in the annual harvest frequency resulted from the greater productive potential of the plant, with more pronounced gains of biomass in the management with annual harvest compared to the biennal harvest, associated with the shorter growth period in the annual harvest.The variation in PRTOC between harvest frequencies resulted from the balance between biomass accumulation and growth time at each harvest interval.
FPR and HFPR were higher in the annual harvest in all combinations of N and P (Table 6).This response is justified by the higher yield of the crop in terms of growth and accumulation of biomass during the shortest development period, a determining condition for the highest production rates of forage and harvestable forage.
In Tejuçuoca, PRFOC was higher in the annual harvesting frequency for most combinations of N and P. The order of superiority in PRSOC varied between annual and biennal harvests, depending on the combination of N and P (Table 7).
The highest PRTOC was observed in the biennal harvest at 77.8% of the N and P combinations.The PRFoOC was higher in the biennal harvest for all N and P combinations.In the N and P combinations that allowed the emergence of fifth order cladodes, higher PRFiOC was verified in the biennal harvest (Table 7).
The highest first order cladode production rate in the management with annual harvest frequency can be attributed to the greater cladode production potential, as a result of the stimulus to the development of buds (areolas) in the matrix cladode, due to the greater incidence of photosynthetically active radiation at the base of the plant throughout the growth period, since the annual harvest provided the removal of the biomass above the primary cladodes each year of cultivation.
The highest production rates of cladodes of higher orders (PRTOC, PRFoOC and PRFiOC) are related to the increase in biomass in the upper stratum of the crop under a longer rest period, since the development time was determinant for the emergence and gain of biomass in the upper parts of the plant.
There were higher FPR and HFPR in the biennal harvest for 66.7% and 88.9% of the N and P combinations, respectively (Table 7).These differences in the harvest frequencies, with predominance of higher values in the  In the analysis of rain use efficiency (RUE) by cactus pear in Quixadá and Tejuçuoca, effects were detected for N and P combinations, fitting to a multiple regression model (Table 8).
The increase in RUE with the increase in the combined doses of N and P in the annual and biennal harvesting frequencies for Quixadá and Tejuçuoca, with a higher amount of cactus pear biomass produced per unit of rainfall, can be attributed to the higher final density of plants, compared to the lowest population of plants in the lowest combinations of N and P, by increasing the response potential of the cactus pear in a denser cactus pear plantation, with a better use of rainfall (RAMOS et al., 2011;SALES et al., 2013).In nutritional terms, it is highlighted that macronutrients increase the efficiency of the plant in transforming water into biomass.The positive effect of nutrients on the root system is also mentioned (ZÚÑIGA-  Harvest frequency (HF); rain use efficiency (RUE); significant at 0.1% (***), 1% (**), 5% (*) and 10% ( ∆ ) probability.TARANGO et al., 2009), which favors the greater use of water by the crop.
In Quixadá, a higher EUC was observed in the annual harvesting frequency for all combinations of N and P (Table 9).The highest mean plant densities that contributed to the highest yields of fresh and dry biomass recorded in the annual harvest in Quixadá support the highest values of EUC found under such conditions.
In Tejuçuoca, the highest RUE was found for the biennal harvest for most combinations of N and P (Table 9).This is because the highest values found for the said RUE index are responsible for the pattern of responses revealed by the biomass in the biennal harvest management.It is also worth mentioning the better structure of the plant, exploring more efficiently the resources of the environment, such as water and nutrients.
The improvement in rain use efficiency by cactus pear fertilized under semi-arid conditions is a relevant response because it reflects the ability of the crop to tolerate low rainfall and its uneven distribution.Harvest frequency (HF); means followed by different letters with the same combination of nitrogen and phosphorus are significantly different by Scott-Knott test at 5% probability CONCLUSIONS 1.The rates of biomass production and rain use efficiency in cactus pear are optimized with the balanced fertilization of combined doses of nitrogen and phosphorus.The combination of nitrogen and phosphorus for maximum biological efficiency of biomass production and rain use efficiency in cactus pear varies according to harvest management and cropping region 2. In Quixadá, aiming at greater persistence and perenniality of cactus pear, it is necessary to adopt a management with biennal harvest and fertilization with 114.6 kg ha -1 year -1 N combined with 136.8 kg ha -1 year -1 P 2 O 5 .
3. In Tejuçuoca, for the maximum biological response of cactus pear as to forage production, with greater persistence and perenniality of cactus pear plantation, it is recommended the management with biennal harvest and fertilization with 190.0 kg ha -1 year -1 N combined with 33.4 kg ha -1 year -1 P 2 O 5 .

Figura 1 -
Figura 1 -Daily rainfall in the regions of Quixadá and Tejuçuoca, State of Ceará, in the agricultural years of 2011 and 2012

Table 1 -
Chemical characteristics of soils in the experimental areas of Quixadá and Tejuçuoca

Table 2 -
Physical characteristics of soils in the experimental areas of Quixadá and Tejuçuoca

Table 3 -
Summary of the analysis of variance of biomass flow and rain use efficiency in cactus pear according to nitrogen and phosphorus combinations under two harvest frequencies, in Quixadá and Tejuçuoca SV -Source of variation; BL -Block; NP -Combinations of N and P; HF -harvest frequency; RE -Residuals; CV -Coefficient of variation; DF -Degree of freedom; PRFOC -Production rate of first order cladodes; PRSOC -Production rate of second order cladodes; PRTOC -Production rate of third order cladodes; PRFoOC -Production rate of fourth order cladodes; PRFiOC -Production rate of fifth order cladodes; FPR -Forage production rate; HFPRharvested forage production rate; RUE -Rain use efficiency; ns -non-significant; ***, ** and * -significant at 0.1%, 1% and 5%, respectively, by F-test

Table 4 -
Production rates of different orders of cladodes of cactus pear in response to the combinations of nitrogen and phosphorus, in the regions of Quixadá and Tejuçuoca

Table 5 -
Production rates of forage and harvested forage in cactus pear in response to combinations of nitrogen and phosphorus, in Quixadá and Tejuçuoca

Table 6 -
Biomass production rates of cladodes according to harvesting frequencies, for different combinations of nitrogen and phosphorus in cactus pear, in Quixadá Harvest frequency (HF); means followed by different letters with the same combination of nitrogen and phosphorus are significantly different by Scott-Knott test at 5% probability

Table 7 -
Biomass production rates of cladodes according to harvest frequencies, for different combinations of nitrogen and phosphorus in cactus pear, in Tejuçuoca Biomass flow and water efficiency of cactus pear under different managements in the Brazilian Semiarid

Table 8 -
Rain use efficiency by cactus pear in response to combinations of nitrogen and phosphorus, in Quixadá and Tejuçuoca

Table 9 -
Rain use efficiency by cactus pear according to different harvest frequencies for different combinaitons of nitrogen and phosphorus, in Quixadá and Tejuçuoca