Performance of Forage Cactus Intercropped with Arboreal Legumes and Fertilized with Different Manure Sources

: The use of agricultural practices such as fertilization and intercropping can improve the production of forage cacti. The objective of this study was to evaluate the agronomic characteristics of forage cactus intercropped with leguminous trees and fertilized with different sources of manure in the tropical semiarid region of Brazil. The research was carried out at the Agricultural and Livestock Research Enterprise of Pernambuco State during the period from March 2011 to September 2013. The following cultivation systems were used: (i) Gliricidia sepium (Jacq.) Kunth + forage cactus cv. IPA-Sertania [ Opuntia cochenillifera (L.) Mill]; (ii) Leucaena leucocephala (Lam.) de Wit + forage cactus; and (iii) forage cactus in monoculture. All of these systems were fertilized with different sources of manure (cattle, goat, sheep, and broiler litter). The goat and sheep manure (16.6 and 16.5 Mg DM ha − 1 year − 1 ) provided the least production of cactus in the different cropping systems. Cattle manure provided greater production of forage and wood from legumes (1.4 and 4.3 Mg DM ha − 1 year − 1 ) and cactus (20.9 Mg DM ha − 1 year − 1 ). Gliricidia produced more forage and wood than Leucaena. Total production of the forage cactus-Gliricidia system produced 4.7 and 3.8 Mg DM ha − 1 of biomass and wood in two years, respectively. The production and morphological characteristics of the cactus increased at further distance from the trees (3 m), and the opposite effect was observed for the concentrations of N, p , and K. Thus, cropping systems using forage cactus and tree legumes fertilized with manure represent an option for tropical semiarid


Introduction
Arid, semiarid, and dry sub-humid regions cover 41.3% of the Earth's surface [1].Semiarid regions account for approximately 15% of the Earth's surface area and support 14.4% of the world's population [2].They are characterized by irregular rainfall distribution (annual mean of 200-700 mm) and an aridity index of 0.20-0.50[3].These conditions limit the growth of forage species and decrease livestock production, one of the most important economic activities in the semiarid regions of the world [4].
The forage cactus (Opuntia sp.) is a plant with enormous productive potential and adapted to the edaphoclimatic conditions of the semiarid region.The presence of crassulacean acid metabolism (CAM) provides greater water use efficiency (100-150 kg of water per kg of dry matter) for the forage cactus, which is approximately six times more efficient than legumes and almost three times more efficient than grasses [4].Rocha Filho et al. [5] suggested forage cactus as the main forage to be adopted in the semiarid region due to its high nutritional value and incomparable production of energy and water per area under Agronomy 2022, 12, 1887 2 of 11 dryland conditions.However, the cactus is not a stable source of feed since it has low concentration of crude protein and fiber [6].Thus, integrating forage legumes such as Gliricidia sepium and Leucaena leucocephala in forage cactus systems complements the diets of the animals, increases N availability through biological nitrogen fixation, and deposits litter with a low C/N ratio, improving nutrient cycling and increasing carbon and nitrogen sequestration in the soil [7,8].
Another key management strategy is fertilizer application, as forage cactus removes a large amount of nutrients from the soil.For every 17 Mg ha −1 of dry matter produced, 304 kg of N, 19 kg of p, 421 kg of K, and 465 kg of Ca are exported per hectare [9].Organic fertilizers such as animal manure (e.g., broiler, goat, sheep, pig, and cattle litter) are affordable, nutrient-rich, and help to improve the physical, chemical, and biological characteristics of the soil.Manure from different animal species have different chemical composition and result in different nutrient releasing rates.Saraiva et al. [10] observed that broiler litter decomposes faster and releases greater amounts of N and p than goat, sheep, and cattle manure.Studies showed that plant management (cropping system, fertilization, among others) directly affects forage production and the morphological and chemical characteristics of forage cactus [11,12].
Our hypothesis is that the decomposition rate of different manures in association with forage cactus intercropped with tree legumes provides better agronomic characteristics.Thus, the objective was to evaluate the forage production and morphological and chemical characteristics of forage cactus intercropped with leguminous trees (Leucaena leucocephala and Gliricidia sepium) and fertilized with different manure types (cattle, goat, sheep, and broiler litter).

Description of the Study Area
The study was conducted at the Experimental Station of Caruaru (8 • 14' S and 35 • 55' W), which belongs to the Agronomic Institute of Pernambuco, IPA, located in the municipality of Caruaru, Pernambuco, Brazil.The predominant soil in the experimental area is Regosol [13], and its chemical characteristics are shown in Table 1.The climate of the site is dry and hot semiarid, classified as BSh according to Köppen, with a rainfall volume and average annual temperature of 727 mm/year and 28 • C, respectively.The accumulated rainfall during the experimental period (March 2011 to September 2013) was 1932.6 mm, with 1022.8;356; and 553.8 mm, respectively, for the 2011, 2012, and 2013 (Figure 1).

Experimental Design and Treatment Description
Treatments consisted of different cropping systems (IPA-Sertânia Cactus + Leucaena, IPA-Sertânia Cactus + Gliricidia, and IPA-Sertânia Cactus in monoculture) and different sources of manure (cattle, goat, sheep, and broiler litter), totaling 12 treatments.Treatments were allocated in a randomized block in a split-plot design and four replications.The main plot (30 × 16 m) comprised the different cropping systems (30 × 16 m), and the sub-plot (30 × 4 m) comprised the different manure sources (30 × 4 m; Figure 2).
The experiment was established in March 2011.The natural vegetation in the experimental area was the Caatinga, which is native to the region.Legumes were planted in three double rows spaced 9 × 1 × 0.5 m apart.IPA-Sertânia cactus was planted between the double rows in monoculture, at a spacing of 1 × 0.25 m [12].In intercropped plots, the density of legumes was 4000 plants/ha and the density of cactus was 32,000 plants/ha, while in isolated cactus cultivation the density was 40,000 plants/ha.The experimental area was fertilized annually (February 2012 and 2013) with different manures aiming to supply 200 kg N/ha.The amount applied was corrected based on the dry matter contents of the different manures (Table 2).Manure application was distributed along the rows of forage cactus, excluding the double rows of legumes.Weeds were controlled, when necessary, by hand-weeding to avoid competition for nutrients.

Experimental Design and Treatment Description
Treatments consisted of different cropping systems (IPA-Sertânia Cactus + Leuca IPA-Sertânia Cactus + Gliricidia, and IPA-Sertânia Cactus in monoculture) and diffe sources of manure (cattle, goat, sheep, and broiler litter), totaling 12 treatments.T ments were allocated in a randomized block in a split-plot design and four replicat The main plot (30 × 16 m) comprised the different cropping systems (30 × 16 m), and sub-plot (30 × 4 m) comprised the different manure sources (30 × 4 m; Figure 2).
The experiment was established in March 2011.The natural vegetation in the ex imental area was the Caatinga, which is native to the region.Legumes were plante three double rows spaced 9 × 1 × 0.5 m apart.IPA-Sertânia cactus was planted betw the double rows in monoculture, at a spacing of 1 × 0.25 m [12].In intercropped plots density of legumes was 4000 plants/ha and the density of cactus was 32,000 plant while in isolated cactus cultivation the density was 40,000 plants/ha.The experim area was fertilized annually (February 2012 and 2013) with different manures aimin supply 200 kg N/ha.The amount applied was corrected based on the dry matter con of the different manures (Table 2).Manure application was distributed along the row forage cactus, excluding the double rows of legumes.Weeds were controlled, when essary, by hand-weeding to avoid competition for nutrients.The experiment was established in March 2011.The natural vegetation in the experimental area was the Caatinga, which is native to the region.Legumes were planted in three double rows spaced 9 × 1 × 0.5 m apart.IPA-Sertânia cactus was planted between the double rows in monoculture, at a spacing of 1 × 0.25 m [12].In intercropped plots, the density of legumes was 4000 plants/ha and the density of cactus was 32,000 plants/ha, while in isolated cactus cultivation the density was 40,000 plants/ha.The experimental area was fertilized annually (February 2012 and 2013) with different manures aiming to supply 200 kg N/ha.The amount applied was corrected based on the dry matter contents of the different manures (Table 2).Manure application was distributed along the rows of forage cactus, excluding the double rows of legumes.Weeds were controlled, when necessary, by hand-weeding to avoid competition for nutrients.

Agronomical Observations and Laboratory Analyzes
Forage cactus were harvested manually in 2012 and 2013 (one and two years after planting), preserving the mother cladodes.Morphological characteristics were analyzed during the biennial harvest.In intercropping, plants were evaluated at a distance of 1, 2, 3, and 4 m from the legumes (two plants/distance/subplot).For the forage cactus in monoculture, this corresponded to four plants/subplot.Plant height was measured from the plant base to its highest point.Plant width was measured in the maximum plant width, and plant length was measured in the area of greatest length.Cladode perimeter was measured around the edge of the cladode.Height, width, length, and perimeter were obtained using a measuring tape.Thickness was measured in the middle part of the plant using a caliper.The number of cladodes per plant was counted in randomly selected plants.From these data, the area of individual cladodes (ACe, cm 2 ) was estimated through the equation: ACe = 1.6691 * (1 − exp (0.0243 * PC))/−0.243[14].
The plants were weighed in the field.A composite sample was prepared and pre-dried in an oven at 55 • C until reaching constant weight [15].Dry matter production was calculated based on plant density per subplot and weight of individual plants.Nitrogen (N), phosphorus (p), and potassium (K) concentrations were analyzed according to the methodology described by AOAC [15].
Legume trees were also evaluated for forage (leaf and stems < 5 mm in diameter) and firewood production two years after planting.The harvest was carried out close to the ground in two plants per subplot.Tree height was measured from the base to the apex of the plant using a measuring tape.

Data Handling and Analysis
Analysis of variance was performed using the Proc Mixed procedure of SAS [16].The fixed effects included cropping system, manure sources, and the distances of the double rows of legumes.The blocks were considered random effect.When the F test was significant, treatment means were compared by Tukey's test at 5% probability.The collection distances were submitted to regression analysis using the Proc Reg procedure of SAS.

Results
The biennial production of forage cactus was affected by the interaction between cropping systems and manure sources (p < 0.05) (Table 3).The use of broiler litter increased production (19.4 Mg DM ha −1 year −1 ) of forage cactus in monoculture.In the systems intercropped with Leucaena leucocephala and Gliricidia sepium, the greatest production (20.2 and 21.7 Mg DM ha −1 year −1 ) was observed when using cattle manure.The least production of forage cactus was observed when fertilized with goat and sheep manure in different cropping systems, with an average of 16.4 Mg DM ha −1 year −1 .There was significant effect between harvests (p < 0.05), in which the biennial harvest (4.2 Mg ha −1 year −1 ) had greater production than the annual harvest (2.5 Mg ha −1 year −1 ) (Table 4).Regarding legumes, Gliricidia sepium was responsible for the greatest production of forage and wood compared to Leucaena leucocephala (p < 0.05).Cattle manure increased the production of forage and firewood of the legume trees in relation to other manure sources.The forage production of forage cactus was also affected by the different distances from legume trees (p < 0.05).There was an increase in production the further the forage cactus was planted from the legumes up to 3 m, although a decrease in production was observed when the legumes were at a distance greater than 3 m (Figure 3).
The number and weight of primary and secondary cladodes were affected by the cropping system and organic fertilization (p < 0.05).The highest number and heaviest weight of primary cladodes was observed in the forage cactus in monoculture (6.7 and 0.69 kg) compared to the intercropped system (5.9 and 0.62 kg).This characteristic was also noticed when the cactus was fertilized with cattle and sheep manure (Table 5).The number of secondary cladodes had similar values for different cropping systems (3.06) and manure sources (3.12).Lighter secondary cladodes were observed in intercropping with Gliricidia sepium (0.52 kg) than in the other cultivation systems (0.55 kg).
The regression analysis showed a quadratic effect for area, length, thickness, width, number, perimeter, and weight of primary and secondary cladodes at different distances from legume trees (p < 0.05).It is noted that there is an increase in these morphological characteristics the further the forage cactus was planted from the legumes up to three meters, although a decrease in production was observed when the legumes were at a distance above 3 m (Table 6).
The forage production of forage cactus was also affected by the different dis from legume trees (p < 0.05).There was an increase in production the further the cactus was planted from the legumes up to 3 m, although a decrease in productio observed when the legumes were at a distance greater than 3 m (Figure 3).Area (cm 2 ), length (cm), thickness (cm), width (cm), perimeter (cm), and weight (kg).
The height of forage cactus and legume trees (Table 7) were modified by cropping systems and manure sources (p < 0.05).Forage cactus showed lower height when intercropped with Gliricidia sepium (0.54 m) compared to forage cactus intercropped with Leucaena leucocephala (0.56 m) and in monoculture (0.57 m).Gliricidia sepium showed greater height (3.4 m), as its plants were 15% taller than Leucaena leucocephala (2.9 m).Regarding manure source, the cattle, sheep, and broiler manure increased cactus height, whereas legumes had greater heights with the use of goat manure.Concentrations of N, p, and K in the forage cactus were affected by the interaction between cropping systems and manure sources (p < 0.05) (Table 8).The concentrations of N and p in the forage cactus in monoculture were similar regardless of the manure source, although the highest K concentration was observed when using fertilization with sheep manure.The different cultivation systems obtained similar N and p concentrations when fertilized with sheep and goat manure, respectively.Forage cactus intercropped with Gliricidia sepium showed a lower K concentration in the different sources of manure.
The regression analysis showed a quadratic effect for N, p, and K concentrations at different distances from legume trees (p < 0.05).The highest N, p, and K concentrations (0.91, 0.20 and 2.06 g kg −1 ) were observed near the tree canopy (0 m), with a reduction of these nutrients (0.80, 0.17 and 1.36 g kg −1 ) occurring at further distances (4 m) (Table 9).

Discussion
Cattle manure provided greater production of forage cactus in intercropped systems, also resulting in greater production of forage and firewood from Gliricidia sepium and Leucaena leucocephala trees.This result can be attributed to the fact that cattle manure was applied in greater amounts compared to the other sources (Table 2) due to its lower N content.The greater amount of manure formed a layer in the soil, probably increasing moisture retention and improving soil physical characteristics (porosity and hydraulic conductivity) [17].It is worth mentioning that manure also has other essential nutrients, such as p and K, which may have favored plant growth and development, as well as micronutrients such as Fe and Mo, which act on oxygen transport and nitrogenase activity and assist in biological N fixation [18].
Studies carried out by Barros et al. [19], Salazar-Sosa et al. [20], and Fonseca et al. [21] prove that organic fertilization using cattle manure increases the production of forage cactus.Silva et al. [22] observed greater production of forage cactus with the application of 80 Mg ha −1 two years −1 cattle manure, obtaining 61 and 90 Mg DM ha −1 two years −1 at densities of 20,000 and 40,000 plants ha −1 .Uddin et al. [23] noted a positive correlation between Leucaena leucocephala and Acacia mangium growth and rates of cattle manure (1:1, 2:1, and 3:1 of soil and manure).
The greatest production in forage cactus in monoculture occurred when it was fertilized with broiler litter.This manure contained the highest concentrations of organic matter, N, p, and K (Table 2), promoting rapid mineralization and nutrient availability.Saraiva et al. [10] noted that the broiler litter has the fastest rates of decomposition and release of N and p compared to cattle, goat, and sheep manure.The organic compounds present in the manure help in the greater solubility and diffusion of p in the soil [24].Studies reveal that p from manure penetrates the soil much more deeply than that from mineral fertilizers [25].
Goat and sheep manure provided the least production of forage cactus and similar concentrations of N and p in the different cropping systems.This may be related to a higher C/N ratio in these manures.The chemical composition of manure is closely related to animal diets [26].Although these animals are more selective and with nutrient-rich excreta, the physical form of goat and sheep feces makes it difficult for soil microorganisms to attack them [27].
Forage cactus produced 41% more forage when harvested at two years than at one year after cultivation, which was probably related to the cumulative effect of the two years of fertilization (2012 and 2013) and the greater rainfall in 2013 (Figure 1).The greater amount of water and nutrients favored cactus growth.Vegetative growth is influenced by soil water content, as key physiological and biochemical processes depend on water, such as photosynthesis, respiration, transpiration, and nutrient uptake [28].This result corroborates studies by Edvan et al. [29] and Ramos et al. [30], who reported that the highest production of cactus increased over time as the plant developed, with no losses caused by tissue senescence.
Gliricidia sepium produced 33% more forage, 39% more wood, and was 15% taller than Leucaena leucocephala.This result may be associated with Leucaena leucocephala being susceptible to ant attack, which hampered its initial development [31].Another factor that may have contributed to this result is the morphological differences between the species.Gliricidia sepium has larger and wider leaves, forming denser crowns than Leucaena leucocephala.A study carried out by Edwards et al. [32] also reports greater biomass production from Gliricidia sepium compared to Leucaena leucocephala.The forage cactus intercropped with Gliricidia sepium showed lower height and lighter cladodes, whereas in the monoculture there was a greater number and weight of primary cladodes, evidencing the negative effect of shading on these morphological characteristics.Santos Neto et al. [33] proved that forage cactus cultivated under 30 and 18% shading had lower height and a smaller number of cladodes.These authors explain that plants close to the woody component in integrated systems present greater competition for light, making these environments critical for the proper development of the cultivated plants [34].
In general, the consortium of forage cactus + Gliricidia produces more biomass, which is one of the main limiting factors for livestock farming in tropical regions.Another advantage of this system is the greater production of wood, which is an important source of energy to meet the needs of domestic households of small and medium-sized farmers.The wood is also used to make permanent live fences, an alternative to replace the use of native plants and reduce human pressure on the caatinga.
Production and morphological characteristics (area, length, thickness, width, perimeter, and number) of the primary and secondary cladodes increased as forage cactus was at a greater distance from the trees (3 m), with these variables decreasing at distances greater than 3 m.This fact can be attributed to greater competition for water, light, and nutrients near trees.Shading reduces stomatal opening and photosynthetic activity, impairing cactus development [35,36].Another aspect is the difference in growth and development between the species studied, resulting in greater competitiveness in the use of resources [37].
On the other hand, N, p, and K concentrations decrease as the distance from trees increases.This occurred possibly due to the fact that there is greater litter deposition under the tree canopy, resulting in an increase in these nutrients in the soil [38].In addition, legumes transfer nutrients to the forage cactus, especially N, via root exudates, root death, and nodules.This transfer can be mediated by mycorrhizae through hyphal connections between legumes and grass roots [39].A study carried out by Villegas et al. [40] proves that there is transfer of N between legumes (Arachis pintoi cv.Mani Forrajero and Pueraria phaseoloides cv.Kudzu) and grasses (Urochloa humidicola cv.Tully, U. brizantha cv.Toledo, and U. decumbens cv.Basilisco).

Conclusions
The cactus and Gliricidia cultivation system fertilized with cattle manure provides greater forage and firewood yields, in addition to better cactus morphological characteristics (cladode height, number, and weight).Forage production and morphological characteristics of primary and secondary cladodes increased at further distance (3 m) from tree trunks.However, under the tree canopy the forage cactus had greater concentrations of N, p, and K.
Cropping systems using forage cactus and tree legumes are an option for semiarid regions.It is possible to produce fodder with high digestible energy from cactus and complemented with the protein and fiber from the legumes.Firewood could be another added value for cooking, considering lack of energy in distant rural areas of several dryland regions across the globe.This system is resilient to climate change and could be further adopted in dryland regions of the world.Policy makers should design strategies to advance these integrated systems in semiarid regions to improve resilience of livestock systems.

Figure 3 .
Figure 3. Forage production of the forage cactus IPA Sertânia at different distances from the a legumes.

Figure 3 .
Figure 3. Forage production of the forage cactus IPA Sertânia at different distances from the arboreal legumes.

Table 1 .
Soil chemical characteristics of the experimental area at different depths.

Table 2 .
Chemical characteristics and the amount applied from different sources of manure.

Table 3 .
Productivity of forage cactus clone IPA Sertânia in different cropping systems and manure sources.

Table 4 .
Forage cactus and leguminous production in different cropping systems and manure sources.
Different uppercase (row) and lowercase (column) letters indicate a significant difference (p < 0.05).CV, coefficient of variation.* Production referring to secondary and tertiary cladodes.

Table 5 .
The number and weight of cladode of forage cactus clone IPA Sertânia submitted to different cropping systems and manure sources.
Different lowercase letters (column) indicate significant difference (p < 0.05) by the PDIFF adjusted for Tukey.CV, coefficient of variation.

Table 6 .
Morphological characteristics of primary and secondary cladodes of forage cactus at different distances from legumes.

Table 7 .
Height of forage cactus and legume in different cropping systems and manure sources.

Table 8 .
N, p, and K concentration of forage cactus in different cropping systems and manure sources.

Table 9 .
Nutrient concentrations (N, p, and K) of forage cactus at different distances from legume trees.