Soil and Root System Attributes of Forage Cactus under Different Management Practices in the Brazilian Semiarid

Abstract

Drylands cover 40% of the global surface and house more than 2 billion people. Drought-tolerant crops are becoming more important in these regions, not only to provide food, fodder, and energy, but also to sequester soil organic carbon. This study evaluated soil and root system attributes of forage cactus ‘Orelha de Elefante Mexicana’ (Opuntia stricta Haw.) managed using different agronomic practices in the Brazilian Semiarid. The experiment was established in June 2011 and the design was split-plot in a randomized complete blocks, where the main plot was the different planting density, and the subplots were the factorial arrangement between harvest frequency and harvest intensity. Soil samples were collected at 0 to 10 and 10 to 20 cm depths and response variables included root biomass, soil bulk density (BD), and soil carbon (C) and nitrogen (N) contents and stocks. Sampling occurred in August 2019, but for root biomass and soil BD analysis it also occurred in September 2021. There were no significant effects from management practices on root biomass at 0 to 10 and 10 to 20 cm depth (p > 0.05), with respective averages of 12.45 Mg ha⁻¹ and 6.06 Mg ha⁻¹. Soil BD was similar at 10 to 20 cm depth (p > 0.05) averaging 1.28 g cm⁻³. Soil organic carbon (SOC) stock varied with management and reached almost 100 Mg C ha⁻¹ in the 0 to 20 cm layer, indicating the potential of cactus to store carbon. Plants with a more developed root system are more likely to survive the drought climatic condition; therefore, less dense plantings could result in more resilient plants for drier regions, but could potentially negatively affect biomass productivity per area.

1. Introduction

The world’s cultivated soils have lost between 25 and 75% of their original stock of carbon (C) [1,2] that was released into the atmosphere in the form of carbon dioxide (CO₂), mainly due to unsustainable management practices resulting in soil degradation and increased climate change and its impacts. Root systems are important contributors of soil organic matter (SOM) formation, by depositing recalcitrant compounds. Root system development, however, is affected by management practices, and little is known about this in relation to cactus.

Thus, good management practices can increase soil organic matter (SOM) content and contribute positively to soil C and nitrogen (N) stocks [3]. In drylands, forage cactus can contribute to enhancing soil organic C, considering its growth potential and high water use efficiency [4]. These practices include frequency and harvest intensity [5], planting spacing/density [4], cultivated species/cultivar [6], edaphic and climatic conditions for cultivation [7], and resistance to attack by pests and diseases [8].

From the agronomic point of view, the ‘Orelha de Elefante Mexicana’ (Opuntia stricta Haw.), as this cactus cultivar is named in Brazil, is originally from Mexico and was introduced by the Agronomic Institute of Pernambuco (IPA) in 1996, provided by the University of Chapingo, Mexico [9]. It has been shown to be less demanding for nutrients, more tolerant to water stress conditions, and has also shown greater production of dry matter (DM) per unit area compared to other genotypes [10]. The IPA Sertânia [Opuntia cochenillifera (L.) Mill.], Miúda, and Orelha de Elefante Mexicana clones are resistant to the carmine cochineal (Dactylopius sp.), and their cultivation areas are expanding in the Brazilian northeast [11].

Santos et al. [12] assessed cactus (O. stricta) productivity at low (6945 plants ha⁻¹), medium (13,889 plants ha⁻¹), and high population density (55,556 plants ha⁻¹). They reported that the productivity of a cactus population equivalent to 13,889 plants ha⁻¹ (medium density) was similar to the productivity using greater populations. They associated this result with the fact that less dense populations need lesser number of cladodes to establish the cactus orchard, reducing costs and facilitating cropping practices. The harvest frequency is also an important management aspect, as it affects the morphological characteristics of cladodes and plant structure [13]. As for harvest intensity, the preservation of a larger residual cladode area promotes more vigorous regrowth and greater longevity of the cactus orchard [14].

However, aspects of the soil and root system of the forage cactus Orelha de Elefante Mexicana need to be further evaluated to assess the potential of cactus to sequester carbon in belowground tissues and SOM formation. We hypothesized that frequent and intensive harvests would affect soil organic carbon, and these harvesting management would interact with planting density. Therefore, the objective of this research was to evaluate soil and root system attributes of forage cactus (Opuntia stricta Haw. Cv. ‘Orelha de Elefante Mexicana’) managed at different planting density, frequency, and harvest intensity in the Brazilian Semiarid.

2. Materials and Methods

2.1. Site Description and Establishment of the Experiment

The field experiment was carried out at the Arcoverde Experimental Station of Agronomic Institute of Pernambuco (IPA) (08°25′ S, 37°04′ W, 681 m altitude) in the semiarid region of Pernambuco, Brazil, with an average maximum temperature of 29.5 ± 2.6 °C and average minimum of 18.5 ± 1.3 °C, and average annual rainfall of 650 mm [15]. The predominant soil is classified as Regolithic Neosol [16]. The cultivar evaluated was ‘Orelha de Elefante Mexicana’ (Opuntia stricta Haw.) and the experiment was established in June (rainy season) 2011.

Plots were laid out as split-plot in a randomized complete block design, with four blocks. The main plot was formed by different planting densities (1.80 × 0.10 m; 1.80 × 0.20 m; 1.80 × 0.40 m; 1.80 × 0.80 m, corresponding to approx. 55,555; 27,777; 13,888; and 6944 plants ha⁻¹, respectively), and the subplots, by the factorial arrangement between harvest frequency (annual and biennial harvest), and harvest intensity (preservation of mother or primary cladodes after harvest). Mother cladodes are the cladodes with a portion inserted into the ground during the establishment, and primary cladodes are the first sprout coming out of the mother cladodes. Second-order cladodes are sprouts coming from the primary cladodes, and so on.

Each main plot measured 10.8 × 8.0 m, each subplot 5.4 × 4 m. In each subplot, three rows of cactus were established with a spacing of 1.8 m between rows. For plots with planting density of 1.8 × 0.8 m, the useful area of sampling corresponded to 4.32 m² and for the other planting densities, it corresponded to 5.76 m².

Annual fertilization of 20 Mg OM ha⁻¹ of cattle manure (OM = organic matter basis) was carried out at planting and harvesting, from the IPA Experimental Station and 200 kg N split into two equal applications, the first at the beginning of the rainy season and the second in the middle of the same period. The source of mineral nitrogen used was urea CO(NH₂)₂, (45% N). Although the N fertilization level applied is apparently high for a semiarid region, it is compatible with the extraction of nutrients by dense cactus cultivation in the semiarid region of Pernambuco [17]. Manure application occurred between rows of the cultivation of the forage cactus Orelha de Elefante Mexicana and on the soil surface. After planting the cactus, weeds were controlled by hand hoe, when necessary.

2.2. Sample Collection and Processing

Samples were collected in each experimental unit, two between cactus rows and two between cacti of the same row at 0 to 10 and 10 to 20 cm depths. Soil physical analyses included bulk density, particle density texture, and flocculation degree. Samples were collected using a “Dutch” auger and composite samples were made from each soil layer and block of the experimental area, and from each layer and each of four representative locations of the native vegetation to characterize the experimental area and native vegetation. In the native vegetation area, there were tree/shrub species as reference.

For determination of soil C and N content and stocks, samples were also collected using a “Dutch” auger. After collection, soil samples were properly stored, identified, air-dried, crushed, homogenized, and processed in sieves with a sieve mesh of 2 mm to obtain air-dried fine earth (ADFE). The samples intended for soil physical analyses (characterization of the experimental area and the native vegetation) were sent to the laboratory of soil physics at IPA, Recife, and those used for soil C and N content and stocks analyses were transported to the Soil and Plant Laboratory of the North Florida Research and Education Center, University of Florida, Marianna, FL, United States of America.

For root biomass and soil bulk density (BD) analyses, samples were collected with the use of a sampler with volumetric cylinder coupled, placed in containers, identified, and sent to the IPA laboratory, Arcoverde. Sample collections for experimental analyses were carried out in August 2019, but for root biomass and soil BD analyses, they were repeated in September 2021.

2.3. Laboratory Analysis

2.3.1. Plant Root Analysis

From the mass of soil with roots and the density of the soil at different depths, the volume of soil collected was calculated, thus allowing the calculation of proportional amounts of mass of dry roots extracted per volume of soil sampled. The roots were separated from the soil by washing with a water jet, in sieves, according to Böhm [18], and taken to drying in an oven with forced air circulation at 70 °C until the samples reached constant mass. Then, the roots were separated from the soil with use of tweezers and weighed on an analytical balance, determining the dry root mass. Values were converted into hectares (Mg of dry matter [DM] ha⁻¹) using the soil BD and the soil mass.

2.3.2. Soil Analysis

The granulometric composition and flocculation degree, as well as the soil BD, determined by the beaker method (characterization of the experimental area) and through the volumetric cylinder method, were determined according to the EMBRAPA methodologies [19].

Soil total C and N contents were obtained through dry combustion (CHNS) in an elemental analyzer. Additionally, soil C and N stocks were determined according to Bernoux et al. [20] and their values corrected by the fixed mass method, according to Sisti et al. [21].

2.4. Statistical Analysis

Physical characterization data of the experimental area and native vegetation were analyzed by comparing sample means and standard deviations.

Root data, soil BD, and soil C and N contents and stocks were analyzed using Proc GLIMMIX of SAS 9.4 (SAS Inst., Cary, NC, USA). For roots and soil BD data, planting density, harvest frequency and harvest intensity, and year were considered fixed effects. The block and its interactions with the fixed effects were considered random. Years were considered repeated measures. As for soil C and N contents and stocks, the planting density, harvest frequency and harvest intensity were considered fixed effects. The block and its interactions with the fixed effects were random. Least squares means (LS means) were considered statistically different at the level of p < 0.05 according to the piecewise differentiable (PDIFF) procedure adjusted by the Tukey test.

3. Results and Discussion

3.1. Root Biomass

Cactus root biomass per area (kg ha⁻¹) was not affected by significant effects of planting density, harvest frequency, and harvest intensity at 0 to 10 and 10 to 20 cm depths (p > 0.05). However, root dry mass per plant (kg plant⁻¹) was significantly affected by plant population in both soil layers (p < 0.05). Thus, stands with greater plant density resulted in individual plants with less root mass. A dense population can lead to greater plant competition, decreasing the growth rate of individual plants [22]. This is relevant given the periods of severe droughts that occur cyclically in the semiarid region. Possibly, plants with more developed root systems are more likely to survive that climatic condition. Larger root systems will also enhance the ability of cactus to use water. Even small rainfall can be efficiently utilized by forage cactus [23]. This is due to the superficiality of the cactus root system, with horizontal distribution in the soil [24], which has the role of absorbing water from light rains and even dew, associated with the high hydraulic conductivity of its roots and the existence of water storage parenchyma tissue in forage cactus [25]. These characteristics of the cactus root system are an advantage for semiarid regions [26], since in addition to this forage being a forage with great use in ruminant feed supplementation, it can be a unique source of water in animal feed during the dry season in these regions [27].

3.2. Soil Responses

Soil texture was predominantly sandy in the experimental area and in the native vegetation at both soil layers (

Table 1. Mean ± sampling standard deviation of physical characteristics of soil samples, according to soil layers (cm) of the experimental area and native forest (NF), Arcoverde-PE.

Local	Soil BD (g cm−3)		Soil Texture (%)
	Apparent	Real	Coarse Sand	Fine Sand	Silt	Clay	FD (%)	TC
Exp. (0–10 cm)	1.2 ± 0.1	2.5 ± 0.1	34.7 ± 2.7	28 ± 2.4	25.7 ± 1.3	11.5 ± 1	96.2 ± 7.5	SL
Exp. (10–20 cm)	1.3 ± 0.1	2.5 ± 0.1	36.5 ± 1.3	28.5 ± 1	22 ± 2.4	13 ± 1.6	85.7 ± 11.1	SL
NF (0–10 cm)	1.3 ± 0.1	2.5 ± 0	47.2 ± 2.9	20.5 ± 1.3	14.2 ± 3	18 ± 2	74.7 ± 2.5	SL
NF (10–20 cm)	1.3 ± 0.1	2.5 ± 0.01	49.7 ± 4.6	18 ± 0.8	13.7 ± 3.4	18.5 ± 1	70 ± 4	SL

Exp. = experiment; NF = native forest; BD = bulk density; FD = flocculation degree; TC = texture class; SL = sandy loam.

). From that perspective, the fractions with the greatest potential for erosion are the silt and sand fractions, especially fine sand, as they do not have a high aggregation capacity, the opposite with clay, which is the most difficult fraction to remove, as it has a greater capacity to aggregation. However, the flocculation degree (FD) averages were ≥70%, values above the optimum of 50% recommended by Goedert [28], a value that suggests reduced susceptibility to soil erosion.

The textural class observed in the soils of the experimental area and native vegetation was sandy loam (Table 1), that is, a loam soil in which sand particles predominate. Loose textured soils, also known as medium textured soils, are defined as soils with similar proportions of sand, silt, and clay particles, which results in good drainage and water retention capacity and an average erodibility index (depending on the degree of flocculation). Sandy loam soils tend to be deficient in both OM and phosphorus, and typically that type of soil needs less care during its management when compared to sandy textured soils [29].

The average soil bulk density for sandy loam soils is 1.51 g cm⁻³ based on the database for Brazilian soils (HYBRAS) and the World Reference Base for Soil Resources (WRB) [30,31]. There were lower soil BD in the cactus cultivation area managed with the preservation of primary cladode in 2019, but in the year 2021, there were no significant differences (p > 0.05) between harvest intensities in the cactus orchard nor in each harvest intensity between the years of evaluation (p > 0.05), and at 0 to 10 cm depth (

Table 2. Interaction of harvest intensity x sampling year for soil bulk density (BD, g cm⁻³) cultivated with forage cactus Orelha de Elefante Mexicana at 0 to 10 cm depth, Arcoverde-PE.

Factor	BD
	g cm−3
	Sampling year
Harvest intensity	2019	2021
Preservation of mother cladode	1.19 Aa	1.12 Aa
Preservation of primary cladode	1.08 Ba	1.11 Aa
p-value	0.03
Standard error	0.02

Different uppercase letters in the column and different lowercase letters in the row indicate a significant difference (p < 0.05).

).

Santos et al. [32] found that cactus cultivar Orelha de Elefante Mexicana was one of the most productive cultivars (55 Mg DM ha⁻¹ year⁻¹), being harvested every other year, conserving the primary cladodes, and under rainfed conditions. This management is important to help plants survive climatic adversities.

There were no significant effects in relation to soil BD at 10 to 20 cm depth (p > 0.05). The soil BD obtained both from the volumetric cylinder method and via the beaker method (characterization of the experimental area, Table 1) were below the critical limit in all treatments and depths evaluated. According to Reichert et al. [33], this critical limit for sandy loam soils (Table 1) is 1.70 to 1.80 g cm⁻³. This upper limit indicates the range of soil bulk density naturally occurring in this soil type, and our values are within this range.

In the 0 to 10 cm soil layer, there was little effect of the management practices on both soil C and N concentrations and stocks (

Table 3. Interaction of planting density x harvest intensity for soil C and N contents (g kg⁻¹) cultivated with forage cactus Orelha de Elefante Mexicana at 0 to 10 cm depth, Arcoverde-PE.

Factor	Soil C Content
	g kg−1
	Harvest Intensity
Planting density	Preservation of mother cladode	Preservation of primary cladode
(plants ha−1)
6944	42.52 Aa	46.44 Aa
13,888	38.28 Aa	57.47 Aa
27,777	45.88 Aa	42.24 Aa
55,555	30.1 Ab	54.7 Aa
p-value	0.01
Standard error	4.52
Factor	Soil C stock
	Mg ha−1
	Harvest intensity
Planting density	Preservation of mother cladode	Preservation of primary cladode
(plants ha−1)
6944	51.4 Aa	50.6 Aa
13,888	46.5 Aa	61.3 Aa
27,777	53.8 Aa	46.6 Aa
55,555	35.5 Aa	56.7 Aa
p-value	0.03
Standard error	5.25
Factor	Soil N content
	g kg−1
	Harvest intensity
Planting density	Preservation of mother cladode	Preservation of primary cladode
(plants ha−1)
6944	3.95 Aa	4.28 Aa
13,888	3.69 Aa	5.45 Aa
27,777	4.35 Aa	3.98 Aa
55,555	2.8 Ab	5.12 Aa
p-value	0.01
Standard error	0.44
Factor	Soil N stock
	Mg ha−1
	Harvest intensity
Planting density	Preservation of mother cladode	Preservation of primary cladode
(plants ha−1)
6944	4.8 Aa	4.7 Aa
13,888	4.5 Aa	5.8 Aa
27,777	5.1 Aa	4.4 Aa
55,555	3.3 Aa	5.3 Aa
p-value	0.03
Standard error	0.53

Different uppercase letters in the column and different lowercase letters in the row indicate a significant difference (p < 0.05).

). The only exception was at the highest plant population density and more lenient harvest, both soil C content and stock, and N content, were greater. In the 10 to 20 cm soil layer, more differences occurred, with a trend for a greater plant population and more lenient harvest, resulting in greater concentrations and stock of C and N (

Table 4. Interaction of planting density x harvest intensity for soil C and N contents (g kg⁻¹) and stocks (Mg ha⁻¹) cultivated with forage cactus Orelha de Elefante Mexicana at 10 to 20 cm depth, Arcoverde-PE.

Factor	Soil C Content
	g kg−1
	Harvest intensity
Planting density	Preservation of mother cladode	Preservation of primary cladode
(plants ha−1)
6944	22.07 ABa	21 Aa
13,888	19.25 ABa	29.38 Aa
27,777	32.86 Aa	21.19 Aa
55,555	18.54 Ba	24.15 Aa
p-value	0.001
Standard error	2.96
Factor	Soil C stock
	Mg ha−1
	Harvest intensity
Planting density	Preservation of mother cladode	Preservation of primary cladode
(plants ha−1)
6944	28.1 ABa	25.4 Aa
13,888	26.1 ABa	36.9 Aa
27,777	42.5 Aa	26.3 Ab
55,555	23.3 Ba	28.6 Aa
p-value	0.003
Standard error	3.81
Factor	Soil N content
	g kg−1
	Harvest Intensity
Planting density	Preservation of mother cladode	Preservation of primary cladode
(plants ha−1)
6944	1.96 ABa	1.85 Aa
13,888	1.79 Ba	2.62 Aa
27,777	3.04 Aa	1.92 Ab
55,555	1.61 Ba	2.24 Aa
p-Value	0.0008
Standard error	0.26
Factor	Soil N stock
	Mg ha−1
	Harvest intensity
Planting density	Preservation of mother cladode	Preservation of primary cladode
(plants ha−1)
6944	2.5 ABa	2.2 Aa
13,888	2.4 ABa	3.3 Aa
27,777	3.9 Aa	2.4 Ab
55,555	2.0 Ba	2.7 Aa
p-value	0.002
Standard error	0.34

Different uppercase letters in the column and different lowercase letters in the row indicate a significant difference (p < 0.05).

). According to Alves et al. [5], maintaining a greater amount of forage cactus cladodes at harvest allows for a greater cladode area index (CAI) remaining after harvesting, which allows for greater photosynthetic efficiency of plants and helps to store C in the soil, since one of the key processes for adding carbon to soil is photosynthesis, in which CO₂ is combined with water, using solar energy to form carbohydrates.

Pereira et al. [34] showed the effect of harvest intensity on the structural development of the forage cactus Orelha de Elefante Mexicana, in which less intense harvest, preserving second-order cladodes, provided greater plant height and width, as well as a greater total number of plants cladodes. Preservation of cladode area allows for greater assimilation of CO₂, increasing the production of new cladodes. These authors found that the management with preservation of primary cladodes contributed to higher forage yield and forage accumulation rate. The effect of plant density likely occurred because at high planting density, the plants are closer together in the crop rows, allowing less solar radiation to reach the lower portion of the canopy [35].

4. Conclusions

Management practices (i.e., planting density, frequency, and harvest intensity) did not affect root biomass per area of forage cactus Orelha de Elefante Mexicana at 0 to 10 and 10 to 20 cm depths. However, greater planting density resulted in lesser dry mass of roots per plant^.. Densely cultivated cactus might result in plants with less roots, individually, and that might decrease the ability of these plants to survive during extreme droughts. Cactus stored approximately 100 Mg C ha⁻¹ in the 0 to 20 cm soil layer, indicating the potential of this crop to help mitigate climate change by storing soil organic carbon in drylands.