Intercropping forage cactus with sorghum affects the morphophysiology and phenology of forage cactus

The aim of this study was to evaluate morphophysiological indices, phenology and cutting time of forage cactus (Opuntia and Nopalea spp.) clones intercropped with sorghum cultivars (Sorghum bicolor). The experiment was carried out from 2017 to 2018, in two cropping systems (monocropping and intercropping), comprising 12 treatments, consisted of three cactus clones: ‘IPA Sertânia’—IPA, ‘Miúda’—Miu and ‘Orelha de Elefante Mexicana’— OEM, in a monocropping system and nine combinations the cacti and sorghum (cultivars 467, SF11 and 2502) to comprise each of the intercropping systems. Cladodes morphophysiological indices were obtained: absolute growth rate, relative growth rate, net assimilation rate (NAR), specific cladode area rate (SCAR) and the cladode area index rate. OEM showed higher absolute growth rates at the beginning of the cycle, compared with IPA and Miu. NAR was higher in the monocropped OEM and in the OEM-2502 intercrop than under all the IPA and Miu systems (monocropping and intercropping), but the SCAR rate was higher for IPA (monocrop and intercropped). Miu showed a shorter duration of phenophase 2, but OEM had a higher rate of cladode emission during this phase. Cactus-sorghum intercropping systems cause a significant reduction in growth rates of the forage cactus. Intercropping has an earlier cutting time than the monocropping.

assimilation and specific leaf area) and trace the physiological and adaptive characteristics of plants (Queiroz et al. 2015;Silva et al. 2019;Souza et al. 2020). These indices help explain crop response to the production environment, favouring more efficient land use and reducing climate risk (Conesa et al. 2017;Xiang et al. 2019).
In the literature, studies using multiple cactus-sorghum cropping systems are still in the early stages, but are in demand, because of the importance of cactus and sorghum as forage (mainly used for animal feed as naturally supplied through grazing/cutting, or as processed silage) in environments with water restrictions (Diniz et al. 2017;Perazzo et al. 2017;Macêdo et al. 2018;Carvalho et al. 2020;Edvan et al. 2020;Jardim et al. 2021a). Under rain-fed conditions cactus (e.g. Nopalea cochenillifera (L.) Salm-Dyck and Opuntia stricta (Haw.) Haw.) is traditionally harvested after two years (Silva et al. 2015a), causing an increase in vulnerability to biotic and abiotic factors. Under irrigated and intercropped systems, cutting forage cactus can be anticipated chronologically, because of changes in growth dynamics that affect the duration of the phenological stages (Amorim et al. 2017;Lima et al. 2018a). It was therefore hypothesised that intercropping cactus with sorghum will alter the growth and development dynamics of the forage cactus, depending on the type of clone used. Furthermore, we hypothesise that it will become necessary to adjust cutting time of the crop, based on the species under cultivation. As such, the aim of this study was to analyse morphophysiological indices, the phenology and cutting time of species of cacti (Opuntia sp. and Nopalea sp.) to improve crop management when these are intercropped with sorghum cultivars.

Description of the experimental site
The experiment was conducted at the Academic Unit of Serra Talhada of the Federal Rural University of Pernambuco, in the municipality of Serra Talhada, Pernambuco, Brazil (7°56′20″ S, 38°17′31″ W and 499 m asl), from 18 March 2017 to 16 June 2018 (a total of 455 days, ~15 months). According to the Köppen classification, the climate in the region is type BSh, i.e. hot semi-arid climate (Alvares et al. 2013), with an average rainfall of 642 mm year −1 and a reference evapotranspiration (ETo) of 1 800 mm year −1 , resulting in a deficit of 1 143 mm year −1 (Silva et al. 2015a). During the experimental period, rainfall was concentrated in February to July 2017 and during the period February to May 2018 (Figure 1), with a total of 997 mm over the two years. The dry period was from August to January. The total water supply by irrigation was 623 mm (1.4 mm day −1 ). The average ETo was 4.71 mm day −1 during the experimental period (i.e. with a maximum of 7.53 mm day −1 and a minimum of 0.40 mm day −1 ), for a total atmospheric demand of 2 431 mm over ~15 months.

Species under study and experimental design
The forage cactus clones 'IPA Sertânia' (IPA) [N. cochenillifera (L.) Salm-Dyck], 'Miúda' (Miu) [N. cochenillifera (L.) Salm-Dyck] and the 'Orelha de Elefante Mexicana' (OEM) [O. stricta (Haw.) Haw.] were used in monocrop systems and intercropped with three sorghum (S. bicolor (L.) Moench) cultivars: 467, SF11 and 2502. These sorghum cultivars and forage cacti were used, because of their rusticity, growth rates and high forage yield (Diniz et al. 2017;Lima et al. 2018a;Jardim et al. 2021a). Before planting, soils were ploughed, harrowed and furrowed, following local practices. In February 2016, the cactus clones were planted at a spacing of 1.0 × 0.2 m (50 000 plants ha −1 ). Until 6 January 2017, plants were grown under rain-fed conditions. To leave the plants uniform, they were harvest on 18 March 2017, keeping the basal cladodes (planted cladodes) and first-order cladodes (arising from the basal cladodes), through a uniformity cut. Irrigation was then carried out, based on the crop evapotranspiration of the forage cactus, following the recommendations of Queiroz et al. (2016). The sorghum cultivars were sown manually on 20 March 2017 in 5 cm deep furrows (using a traditional hand hoe and covered with a thin layer of soil). These cultivars were sown at a lateral distance of 25 cm from each row of the forage cactus (i.e., each crop sequence consisted of four planting rows). Twenty-eight days after seedling emergence, approximately 20 plants per linear metre remained (200 000 plants ha −1 ). A timeline of the key experimental procedures is presented in Supplemenatry Figure S1 (for more details see supplemental material).
The experiment had a randomised block design (RBD) and 12 treatments, three forage cactus clones in monocrop systems (i.e. IPA, Miu and OEM) and nine intercropped systems, based on the maximum number of combinations between the forage cactus clones and the sorghum cultivars (467, SF11 and 2502), arranged as follows: IPA-467, IPA-SF11, IPA-2502, Miu-467, Miu-SF11, Miu-2502, OEM-467, OEM-SF11 and OEM-2502. The treatments consisted of four plots (replications), giving a total of 48 experimental units consisting of four rows (of 25 cactus plants), with an area of 60 m 2 . Each measurement plot consisted of the two central rows, with 46 working plants (the plants at each end were also discarded) and a total area of 16.4 m 2 .

Irrigation management and cultural practices
In the current study, the irrigation was by means of a drip system at a flow rate of 1.75 ± 0.14 l h −1 and a working pressure of 101.32 kPa, using water from an artesian well located near the experimental area, with a mean electrical conductivity of 1.51 dS m −1 (classified as C3, i.e. as high-salinity water) according to Richards (1954): pH = 6.84, Na + = 168.66 mg l −1 and K + = 28.17 mg l −1 . Although we have intercropping systems, the irrigation was based on the water requirement of the forage cactus crop (i.e. because it is the main crop), using a crop coefficient of 0.52 and the reference evapotranspiration (ETo) (Queiroz et al. 2016). The ETo was calculated using meteorological data collected daily from an automatic weather station belonging to the National Institute of Meteorology. To determine the ETo, the Penman-Monteith method was used (Allen et al. 1998).
During the experiment, mineral fertiliser was applied twice throughout the area, the first application in January and the second in August 2017, in the formulation 14-00-18 + 16 S, using 525 kg ha −1 (73.5 kg N ha −1 , 94.5 kg K 2 O ha −1 and 84 kg S ha −1 ), as recommended by Diniz et al. (2017). All the crop basic health and maintenance treatments in the experimental area (cleaning weed and the control of pests and diseases) were carried out throughout the period of crop development. Because of the use of irrigation, the high soil moisture promoted the proliferation of the Fusarium solani (Mart.) Sacc. and Erwinia carotovora subsp. carotovora Jones, which are soil-borne plant pathogens. For this reason, the mortality rate of the plants was 81.62% ± 1.45 (IPA), 71.68% ± 3.71 (Miu) and 8.56% ± 1.29 (OEM); this type of characteristic is also being reported by other researchers (Silva et al. 2015b;Araújo Júnior et al. 2021;Jardim et al. 2021a) and generally cause cladode soft rot, necrosis and high mortality (Swart 2009).
One cactus cycle of 15 months and four consecutive sorghum cycles were monitored. During the first cycle, the sorghum was harvested 120 days after emergence; during the second cycle, cultivar 2502 was harvested 71 days after cutting (DAC) and cultivars 467 and SF11 at 92 DAC. For the third cycle, cultivar 2502 was harvested at 94 days after the second cutting and cultivars 467 and SF11 at 102 days after the second cutting; during the fourth cycle, cultivar 2502 was cut at 84 days after the third cutting and cultivars 467 and SF11 at 85 days after the third cutting. All sorghum cultivars were harvested when the plants were with the soft dough stage (Vanderlip and Reeves 1972).

Evaluation of the growth, phenology and cutting time of the forage cactus
Morphometric measurements of the cactus plants were taken monthly and every 90 days biomass data were collected to quantify the morphophysiological indices, phenology and cutting time. Morphometric measurements were taken in two plants per plot, where the number of cladodes (NC, units) was counted in order of appearance on the plant (first-order, second-order and so forth); in addition, the length (cm), width (cm) and perimeter (cm) of the cladode was measured, based on the procedures of Jardim et al. (2020a). The cladode area (CA, m 2 ) was calculated from the morphometric data, which, with the spacing between plants, was used to estimate the cladode area index (CAI, m 2 m −2 ). The cladode area was calculated using Equations 1, 2 and 3 for each of the forage cactus clones and the cladode area index was calculated by the where, CL is the cladode length (cm); CW is the cladode width (cm); CP is the cladode perimeter (cm); i is the observation number; n is the total number of observations; 10 000 is the conversion factor from cm 2 to m 2 ; and S1 × S2 are the spacing between the rows and plants of each forage cactus clones (i.e. 1.0 × 0.2 m), respectively. The cladode dry matter of the clones (W, Mg ha −1 ) was obtained by harvesting the plants (one plant replica per plot, i.e. four repetitions in total) over seven dates (biomass evaluations): at 90 days intervals, starting on 18 March 2017 and last analysed on 16 June 2018. The plants were weighed (to obtain the fresh weight) with the aid of an electronic balance. Two cladodes from the middle third of the plant were then broken up and packed in paper bags. The cladodes were left in a forced air ventilation oven at 55 °C to constant weight (approximately 10 days, obtaining the dry matter) (Jardim et al. 2021b).
Using the morphometric and dry matter data, the morphophysiological indices and the phenology were calculated for the cactus, adjusting three-parameter sigmoid models (Equation 5) and using the accumulated degree-days (ADD) as an independent variable, because this method is applied more in phenological evaluations, because of the crop responses to environmental conditions (McMaster and Wilhelm 1997). The ADD was obtained with Equation 6, proposed by Arnold (1959).
where, y are the NC, W or CAI (response variables); a is the maximum value for the rate (i.e., distance between the two asymptotes); x is the accumulated degree-days; x0 is the number of degree-days necessary for the plant to express 50% of the maximum rate (i.e., inflection point of the curve); and b is the number of degree-days necessary for the start of the rate. The sigmoid model is chosen, based on the significance of the model (p < 0.05; using the F-test) and coefficient of determination greater than 0.85.
where, ADD is the accumulated degree-days (°Cday); i is the daily time step; n is the total number of days; Tmax is the maximum daily air temperature (°C); Tmin is the minimum daily air temperature (°C); and Tb -is the lower base temperature (°C). The lower base temperature used was 22 °C, as per an earlier analysis (Souza et al. 2020).
Negative ADD values were set to zero. The ADD is a type of heat units, particularly used for crop phenology studies with high precision (McMaster and Wilhelm 1997). From the derivative of Equation 5 (i.e. to describe the new Equation 7) and the adjusted parameters for the cladode dry matter and the cladode area index, the morphophysiological indices were obtained the absolute growth rate (AGR, Mg ha −1 °Cday −1 ), relative growth rate (RGR, Mg Mg −1 °Cday −1 ), net assimilation rate (NAR, Mg ha −1 °Cday −1 ), specific cladode area rate (SCAR, ha Mg −1 ) and the cladode area index rate (CAIR, ha ha −1 °Cday −1 ). For the phenology, the parameters used were adjusted by the number of cladodes in order of appearance on the plant (NC1 is number of first-order cladodes, NC2 is number of second-order cladodes, and so forth).
The values for AGR, RGR, NAR and SCAR in the cactus were evaluated over time, as per a method adapted from Ramírez-Valiente et al. (2017), being AGR = ∂W/∂t, RGR = ∂W/∂t/W, NAR = AGR/CAI and SCAR = ∂CAI/∂t/AGR. The procedure defined by Amorim et al. (2017) was used for the phenology of the cactus, where the change in phenological phase of the plant depends on the emission rate of two successive-order cladodes. The cutting time was established when 25% of the peak maximum AGR of the cactus was reached.

Statistical analysis
The mean values for the growth rates of the forage cactus throughout the cycle were compared among treatments. They were submitted to tests of normality (Shapiro-Wilk test), homoscedasticity (Bartlett's test) and also the statistical method one-way analysis of variance (ANOVA) by F-test (p < 0.05) to the treatments applied-i.e. cropping systems. The normality and homoscedasticity tests were applied before ANOVA and all analysed variables presented normal distribution. Finally, all the statistical analysis was carried out using the R software (R Core Team 2021) with 'agricolae' package (Mendiburu 2020). Figure 2a shows  (Figure 2a), mainly for the OEM-SF11 system (see inset in Figure 2a). Through the insets in rate panels, it is possible to have in more detailed visualisation the variation of the rates of the cropping systems at the end of the degree-days. In general, for monocropping systems the Nopalea clones (IPA and Miu), the AGR curves were similar over time (Figure 2a). However, when intercropped with the sorghum cultivars, the AGR was affected (p < 0.05). Miu achieved higher growth rates (0.0046 Mg ha −1 °Cday −1 ) in the monocrop system, compared with IPA. Under intercropping conditions, the highest AGR for the IPA clone occurred in Miu clone, the rates were higher in the intercropping system with Miu-467, followed by the configuration with Miu-SF11 and Miu-2502 (p < 0.05), reaching 0.0092, 0.0073 and 0.0055 Mg ha −1 °Cday −1 , respectively (Figure 2a). Figure 2b shows the variations in RGR between the cactus clones in their respective configurations, with higher values noted at the beginning of the cycle. In the OEM intercrops, the highest values for RGR were seen in the OEM-467 and OEM-SF11 arrangements (p < 0.05), of 0.0088 Mg ha −1 °Cday −1 and 0.0060 Mg ha −1 °Cday −1 , respectively. These results are on average twice as high as those for the configurations with OEM-2502 and for the monocropping system. In the systems with the Nopalea clones, Miu had higher values than those obtained by IPA. For the three clones, there was a sharp reduction at 600 °Cday in RGR, however, for the IPA clone, the reductions were less pronounced.

Growth analysis
The NAR for the three cactus clones is shown in Figure  3a. intercrop stood out, followed by the monocrop. The OEM and Miu clones had the lowest rate for SCAR, compared with the IPA clone ( Figure 3b). IPA had a value for SCAR of the order of 10.97 ha Mg −1 in the monocropping system, followed by the IPA-SF11 intercrop, with 5.54 ha Mg −1 . Figure 4 shows the results for the CAIR for the three forage cactus clones, where OEM had the highest rates, both at the beginning of the cycle and over time (varying from 0.0044 ha ha −1 °Cday −1 to 0.0052 ha ha −1 °Cday −1 ). The intercrops directly affected the rate in the Nopalea clones, with IPA more prominent than Miu; however, the reductions in CAIR in Miu were less pronounced over time. Figure 5 shows the phenophases, the dry matter rate and the cutting time for the forage cactus clones. In Figure 5, only variations in the of phenophase 2 (i.e. phenophase 2.1; is the initial stage of phenophase 2 -to designate the first period with the half greatest emission of secondorder cladodes) and phenophase 2.2, being the final stage of phenophase 2 -to designate the second period with the half greatest emission of second-order cladodes) and phenophase 3 (i.e. phenophase 3.1 and phenophase 3.2, initial and final stages of phenophase 3, respectively) are shown. In the intercropped Nopalea clones, the phenophases were of longer duration than in the monocrops of cacti; this was not seen in most of the Opuntia OEM clones, only in the OEM-467 configuration (Table 1). Miu, regardless of the cropping system, had the shortest duration for phenophase 2 (i.e. the period with the greatest emission of second-order cladodes), compared with the other clones, with a mean duration of 466 °Cday. For IPA, the duration of phenophase 2 was, on average, 766 °Cday, longer than OEM (510 °Cday), however, the OEM clone had the highest rate of cladode emission (e.g. second-order cladodes). The systems showing the longest duration for this phenophase were IPA-467 and IPA-2502 (1 041 °Cday and 903 °Cday, respectively). Miu was the only clone that exhibited phenophase 3 (i.e. the period of greatest emission of third-order cladodes), with a mean duration of 1 454 °Cday; the systems with the longest duration were Miu-467 and Miu-2502 (1 678 and 1 650 °Cday, respectively), whereas Miu-SF11 had the shortest duration (1 490 °Cday).

Phenophase and cutting time of the forage cactus
The effects of intercropping on the duration of the phenophases resulted in a reduction in cutting time in the Nopalea clones. For OEM, the cutting time was reduced in OEM-467 and OEM-2502 in relation to the monocrop, showing that, in general, the cactus-sorghum intercrop has an impact on cutting time in the cactus. The OEM clone displayed an average of 676 °Cday as necessary for harvesting (see Figure 5 and Table 1) and the OEM-467 arrangement required a thermal time (also known as heat units, expressed in accumulated degree-days or growing degree-days) of 405 °Cday for the ideal cutting time, with OEM-2502 needing 658 °Cday. OEM-SF11 presented a longer thermal time (855 °Cday) until cutting. In turn, the monocrop OEM system required 786 °Cday. The IPA clone had the longest cutting time among the cacti, of 1 025.75 °Cday, followed by Miu with 878.25 °Cday. It was found that the monocropped IPA and Miu systems had a similar total thermal time until cutting (1 614 °Cday and 1 584 °Cday, respectively). Among the intercropped systems, IPA-467 needed 1 249 °Cday: longer than IPA-SF11 and IPA-2502, which required 467 °Cday and 773 °Cday respectively. In the Miu intercrops, the thermal time until cutting was 643 °Cday, being shorter for the Miu-467 configuration (514 °Cday), with 641 °Cday and 774 °Cday for Miu-2502 and Miu-SF11, respectively (Table 1)

Growth analysis
The OEM forage cactus, both in the monocrop system and intercropped with sorghum 2502, had the highest rates of dry matter accumulation from the start of the cycle, compared with the IPA and Miu clones in their various configurations. This result is related to the aggressiveness of the OEM clone, especially in the semi-arid environment (Diniz et al. 2017), because this clone, even in adverse conditions, has a high adaptive capacity (Jardim et al. 2021a). According to Ramírez-Valiente et al. (2017), species xerophilic adapted to semi-arid climates express higher growth rates when exposed to favourable environments. Therefore, cacti of the genus Opuntia, when submitted to irrigation, show better adaptation and plasticity of the photosynthetic metabolism (i.e. facultative CAMcrassulacean acid metabolism) and, as a result, display better performance in hostile environments (Winter et al. 2008;Jardim et al. 2021b).
The impact on the AGR of the cactus was very similar in the OEM-467 and OEM-SF11 intercrops, because of these sorghum cultivars being phenotypically alike. The AGR observed in OEM were greater than those reported by Queiroz et al. (2015), at 380 days of age, during the second production cycle and under full irrigation. The monocropped Nopalea clones were similar for biomass accumulation per unit area, but when intercropped, showed variations in AGR depending on the type of sorghum cultivar (467, SF11 and 2502). Miu achieved higher rates, because of its greater capacity for cladode emission, peculiar to this clone (Müller et al. 2018).
A striking feature of the species is the RGR, which explains the variation in AGR (Souza et al. 2020;Araújo Júnior et al. 2021). For example, the competition between OEM and the cultivars 467 and SF11 (taller sorghum) led this clone to display a greater initial growth rate, when sorghum has presented a height lower than 1 m. These factors were observed in intercropping systems with OEM-467 and OEM-SF11, because of the effects of interspecific competition on the growth parameter of the cactus. Similar results were observed for the Nopalea clones, especially Miu, which displayed greater RGR values than did IPA. In addition, IPA showed a less pronounced reduction than did the other two clones; this is related to its growth habit, which is slower and more erect. RGR is attributed to the morphological characteristics of the plants (i.e. leaf mass fraction, leaf area), so variations can occur between species and age. Over the course of the cycle, because of crop growth and self-shading, the RGR was reduced and photoassimilates were likely allocated to other plant structures that were still under development (Müller et al. 2018;Vasseur et al. 2018;Souza et al. 2020;Jardim et al. 2021b).
Similar to that seen for AGR, the monocrop OEM system and the intercrop with OEM-2502 showed the greatest values for NAR among all the configurations and clones. The NAR expresses the carbon gained by a species via net photosynthesis (Huber et al. 2018). The shading of grasses by the canopy reduces overall interception of the light energy, resulting in a reduction in the NAR (Herce et al. 2014). The NAR can be compromised, depending on the position of the cladodes on the plant, as well as the opacity of the surface, because of reflectance and/or blockage of the solar radiation. These reductions in light energy negatively affect the flow of photoassimilates in cacti of the genera Opuntia and Nopalea (Nobel and Zutta 2008).
The OEM and Miu clones showed lower rates for SCAR, compared with the IPA clone; in the intercrops, the SCAR was even lower in relation to the monocrops. The reason for this is the reduced NAR, a result of the intercrop and self-shading of the cladodes, which results in less benefit from photosynthesis (Queiroz et al. 2015). In a study by Ojeda-Pérez et al. (2017) employing Opuntia cacti, the authors found that reductions in SCAR were related to greater water loss from the cladodes or the reallocation of sugars. Because cactus clones typically have low levels of dry matter and OEM and IPA have a larger cladode area, this explains the lower SCAR, compared with that of Miu (see Figure 3b). The OEM clone had the greatest increase in cladode area index; whereas the intercrops affected these rates, which confirms the effect on the SCAR.

Phenophase and cutting time in the forage cactus
The results showed that the Miu clone showed the shortest duration for phenophase 2 among the clones under study, but in OEM, which requires a higher thermal time to advance the phenophase. Furthermore, the highest rate of cladode emission is during this phase, marking the greatest contribution to its productivity. The phenophase 2, is designated with the greatest emission of second-order cladodes of the cacti. Several studies also reported that the greatest contributions of cladodes and biomass to the final yield of OEM are in fact associated with the second-order cladodes (Pinheiro et al. 2014;Amorim et al. 2017). The prolongation or precocity of a phenological phase is related to interspecific competition for light, water and nutrients. López-García et al. (2016) state that the age of the cactus can be seen from the size of the plant or the number of cladodes, but these depend on the environmental conditions and the flow of carbohydrates from the basal cladodes. Amorim et al. (2017), following the phenology of the OEM cactus, found a duration of 6.3 months for phenophase 2 (~550 °Cday). Interaction between the genetics and solar radiation promotes changes in the onset and duration of the phenophases (Motsa et al. 2017), because each cactus has its own thermal and water requirements, which alter the dynamics of its phenology (Gomes et al. 2019).
In cactus clones, understanding the behaviour of this crop is of significant importance. Variations in the duration of phenophase 2 in the IPA clone were longer than that of OEM and Miu, whereas Miu was the only clone that exhibited phenophase 3. This phenophase 3, was the longest among the clones and is a result of the higher emission rate of third-order cladodes; in the Miu-467 and Miu-2502 intercropping systems. Amorim et al. (2017) stated that knowledge of the duration of the phenological stages of the cactus is indispensable for improving management.
The cactus-sorghum intercrop had a direct impact on cutting time in the cactus. The environmental conditions and the genetic factors of the crops are key to understanding phenological performance because they influence the growth rate (McMaster et al. 2011). In the current study, crop configuration was also crucial in modifying the cutting time of the cactus. Competition for natural resources changes photo-assimilate distribution and cladode emission (Bowers 1996), which cause changes in the dynamics of the rates of cactus cladodes, as also morphological and structural changes. Under full water availability, carbon assimilation is maximised and, as a result, also crop productivity. According to Cruz and Pavón (2013), when cacti are submitted to water restrictions, even though they have structures for water storage, their development is restricted. Because cacti were irrigated in this study, the favourable conditions allowed for maximum cladode emission during the experimental period, including Miu, which emitted third-order cladodes delaying its cutting time, compared with the OEM and IPA clones.

Conclusions
The use of cactus-sorghum intercropping systems resulted in variations in the growth dynamics of the forage cactus, affecting its growth. The use of sorghum as an intercrop causes a significant reduction in growth rates. As a result of the intercrop, the Nopalea clones showed greater variation in phenological-phase duration, proving to be more sensitive to this type of system with the various sorghum cultivars. Regardless of the clones, each cactus had at least two intercrop configurations with an earlier cutting time than the monocrops. The systems that included 'Orelha de Elefante Mexicana' and the sorghum cultivars 467 and 2502, resulted in the smallest negative impact on cactus growth. For the Nopalea clones, the greatest benefits came as a result of the higher emission rate of cladodes and cutting time from the 'Miúda'-467 configuration, with the above systems proving to be promising in semi-arid environments. Additional research is needed to verify the factors that inhibit growth, yield performance and phenological development of the forage cactus in intercropping systems under other environmental conditions.