Effect of Pruning in Coffee-Based Agroforestry System on Soil Chemical Properties

The low coffee productivity in smallholder coffee-based agroforestry systems and soil fertility degradation indicate mismanagement in soil and plant systems. The study examined pruning management practices on soil chemical properties in smallholder coffee agroforestry systems. Different pruning management were examined on UB forest coffee-pine agroforestry, namely pruned pine and unpruned coffee (T1), unpruned pine and pruned coffee (T2), and unpruned pine and unpruned coffee (T3). Each treatment has four replications, a total of 12 plots (size 2×3 meters for each plot). Soil samples were taken from five sampling points in each plot, ranged between 0-50 cm from the coffee stems at two different soil depths (ie, 0-20 cm and 20-40 cm) in each plot. Unpruned pine and coffee pruned (T2) plots had 21% to 61% higher total N at 20-40 cm soil depth than the other plots. Meanwhile, total K, soil exchangeable K and Na were higher from 21% to 335% in plots with pine and coffee without pruning (T1) as compared to unpruned pine with pruned coffee (T2) and pine and coffee without pruning (T3), preferably at 0-20 cm soil depth. Lastly, the unpruned pine and unpruned coffee management (T3) had 98% to 162% higher soil exchangeable Ca and Mg at a depth of 0-20 cm as compared to the other plots. The study suggests that tree management can affect soil chemistry. The study also found differences in litter input and microclimate (i.e., temperature, moisture and humidity levels, both on soil and air) between plots which are potentially affected by decomposition rate of soil organic matter and nutrient mineralization.


Introduction
Coffee is one of plants having a high economic value and is commonly cultivated in agroforestry systems.Agroforestry systems combine several service benefits (environmental, economic, and cultural) by providing a complex environmental structure [1].The application of agroforestry in coffee plantations provides better sustainability by reducing adverse impacts and mitigating microclimate fluctuations in the system [2].The presence of shade plants modifies the microclimate to make it more suitable for the living conditions of coffee plants.The presence of shade provides a stable microclimate that has a significant impact on coffee production [3].Additionally, coffee agroforestry with a good management system can maintain soil health.Agroforestry systems reportedly increase litter input that important for earthworms, giving positive effect on coffee beans dry weight [4].Soil organic C, microbial biomass, and exchangeable K were reportedly increases, approaching that of secondary forests [5].The positive impact of implementing agroforestry has made this system sustainable for coffee cultivation.The Universitas Brawijaya Forest (UB forest) is one of the locations that implements coffee-based agroforestry.Many small farmers have applied coffee-based agroforestry systems to UB forests.On average they use pine as stands and do it in an area of 0.7 ha [6].The absence of coffee-based agroforestry land management standards causes high management variability, which encourages mismanagement of coffee-based agroforestry lands.This causes low coffee production in the UB forest and a decrease in soil quality.Improper management increases the chances of soil degradation and reduces the availability of nutrients in the soil [7].Pruning management of trees (e.g., coffee or pine) is one of the factors contributing to the high variability in management.Pruning is necessary to maintain the land conditions.The results of the study reported that coffee fields with above-average shade density caused coffee production to decrease owing to light competition [8].Inappropriate application of coffeebased agroforestry management gives some negative impact, such as declining nutrient availability.Management variability can be reduced by applying the proper management simultaneously, one of which is cut the unproductive branches, both on coffee tree or shading tree.Cutting the unnecessary branches attempts to reduce the number of branches that are unproductive.Coffee pruning and pine trimming are two management practices applied by farmers in the UB forest.The effects of pruning have been reported in several previous studies.Pine pruning increases the intensity of light entering the system and stabilizes soil moisture [6].Pruning unproductive coffee branches helps reduce competition for nutrients, thereby encouraging the optimization of flower formation [9].However, initial information on the effect of different management strategies on soil fertility in the form of pruning coffee agroforestry plants in UB forests has yet to be reported.This study examined the pruning management effects on soil chemical properties under coffee-pine agroforestry systems in UB forests.

Study site and research design
This research was conducted on coffee-based agroforestry land located in the UB Forest, Karangploso District, Malang Regency (Figure 1) , and laboratory analysis was conducted at the Soil Chemistry Laboratory, Soil Science Department, Faculty of Agriculture, Universitas Brawijaya.UB forest is located on the slopes of Mount Arjuno, and the research was taken at an altitude of 1,200 meters above sea level (m asl), with coffee and pine as the plants found in each plot.A randomized block design (RBD) with three different tree management practices, namely: pine-pruned and coffee unpruned (T1), pine-unpruned and coffee-pruned (T2), and pine-unpruned and coffee unpruned (T3) was used on this study.Each management was repeated four times, totaling 12 plots (Figure 2a).Each plot measured 2 m × 2 m.The pines were pruned by pruning branches at a height of 10 m [10].Coffee pruning is performed by pruning unproductive branches, and keeping the coffee plants height around 150 cm.

Soil sampling method and analysis
Two soil depths (i.e., 0-20 cm and 20-40 cm) were sampled to determine soil nutrient status of each management (Figure 2b).Soil samples for each plot were collected at 50 cm from the coffee stems.Four sampling points surround the coffee stems.Soil sampling was conducted using a soil auger.the obtained soil sample was air-dried and sieved through 0.5 mm and 2 mm.
A 2 mm sieved soil samples were used for soil pH analysis, while 0.5 mm sieved soil samples were used for the analysis of soil organic carbon (SOC), total nitrogen (TN), total phosphorus (TP), total potassium (TK), available phosphorus (AP), and exchangeable bases (K, Na, Ca, and Mg).Actual soil pH was measured using a H2O extractant, potential soil pH was measured using a 1N KCl extractant.SOC analysis was performed using the Walkey-Black method.TN analysis was performed using the Kjeldahl method.TP and TK were determined using the 25% HCl extraction method.AP was analyzed using the Bray 1 method.Soil exchangeable base analyses (K, Na, Ca, and Mg) were performed using the NH4OAc pH 7 extraction method.

Data analysis
The Shapiro-Wilk test was used to test the normal distribution of the data.Logarithmic or square roots transformations were used to normalize the data.Normal or transformed data were tested using ANOVA (F-test) at a 95% confidence level.Fisher's Least Significant Difference (LSD) test was carried out to determine differences between tree management practices.

Soil pH
Different pruning management did not affect the actual soil pH at any depth or the potential pH at a 0-20 cm soil depth (p>0.05)(Figure 3).The difference in tree management only affected the potential soil pH at depth of 20-40 cm (p<0.05).
Pine-pruned and coffee unpruned management (T1) had the highest potential pH (4.70 ± 0.02) compared to other treatments.The actual soil pH mostly higher than the potential soil pH value on this study due to KCl can extract more Al 3+ and H + than H2O extractant [11].This ability caused the potential soil pH to tend to be lower than the actual pH.In addition, the lower potential soil pH than the actual soil pH showed a net negative charge in the colloid (cation exchange capacity).The results of this study revealed that T1 has the potential to have a higher alkaline content in soil colloids at depths of 20-40 cm.The low coffee production in the T1 plot compared to other plots resulted in lower transport of nutrients out of the system, which resulted in higher accumulation of Ca 2+ and K + and easier leaching.Excessive amounts of Ca 2+ and K + in soil colloids increase H + exchange process, resulting in a higher potential pH [12].

Soil organic carbon and total soil nitrogen
Different tree management had an effect on soil organic carbon at a depth of 20-40 cm and total soil nitrogen at all depths (p<0.05)(Figure 4).Pine-unpruned and Coffee pruned management (T2) showed the highest SOC at a depth of 20-40 cm (8.79±0.67%C) and the highest TSN at all depths (0.60±0.01%N at 0-20 cm and 0.56±0.004%N at 20-40 cm).The TN at T2 increased from 21% to 61% compared with the other treatments.Similar results have been reported in previous studies, where the presence of litter was positively correlated with SOC and TN [13].Pruning can improve the microclimatic Coffee pruning increasing the accumulation of pine litter, which predominates at ground level.Temperature affects the amount of N released from the soil [16].The presence of pine litter helps suppress N volatilization; therefore, it is not easily removed from the system.In addition, pruning helps transport nutrients to productive branches [17].These conditions caused N to be used in smaller and more abundant amounts in the soil than in the other systems tested.

Soil P and K
The results showed that differences in tree management did not affect TP and AP in the soil at any depth (p>0.05)(Figure 5a and 5b).This was because the parent material formed the UB forest.UB Forest's location on Mount Arjuno's slopes causes the soil forming materials to be rocks and volcanic materials.As much as 74% of the total land in the UB forest identified as andic humudepts subgroup [18].The presence of andic properties in soil causes many humus-aluminum complexes and allophane minerals to bind P in the soil, making them unavailable to plants [19,20].P is an important element in coffee production.This is one of the factors that need to be considered in the management of coffee agroforestry in UB forests so that production can increase.
Different tree management shows an effect on TK and exchangeable K (Figure 5c and 5d).Pine-pruned and coffee unpruned management (T1) showed the highest value for TK (682.41±66.26mg.kg -1 at 0-20 cm, and 589.30±46.71mg.kg -1 at 20-40 cm) and soil exchangeable K (2.06±0.38 me.100g -1 at 0-20 cm, and 1.83±0.21me.100g -1 at 20-40 cm) compared to other management.The amount of coffee produced in each plot is thought to influence this result.Coffee residue has high K and when it is returned to the system, it provides K for plants [21].The low yield in plot T1 causes low K transport out of the system such that the amount of K in plot T1 is stable.

Soil exchangeable base (Ca, Na, Mg)
The different tree management methods resulted in significant differences in exchangeable Na, Ca, and Mg.Pine-pruned and coffee unpruned management (T1) had the highest exchangeable Na (1.40±0.04me.100g -1 at 0-20 cm, and 1.44±0.01me.100g -1 at 20-40 cm), compared to all management at all depths (p<0.05)(Figure 6a).Canopy openness caused by pine trimming is thought to be the cause of high exchangeable Na.The presence of a canopy has an impact on higher intercept values and lower land canopy openness [22].This led to low exchangeable Na values in pruned and unpruned coffee management.
Pine-unpruned and coffee unpruned management (T3) had higher Ca and Mg values compared to other management systems at a soil depth of 0-20 cm (19.77±0.89me.100g -1 and 11.63±1.31me.100g -1 , respectively) (Figure 6b and 6c).The level of nutrient pumping was thought to have influenced these findings.Nutrient pumping is a mechanism by which shade plant roots bring Ca 2+ and Mg 2+ to the topsoil, making it more available to plants [23].Slowing nutrient cycles, high shade tree density, and low decomposition processes are impacts of coffee agroforestry management without pruning [24].The slowing down of nutrient cycles encourages more massive nutrient pumping compared to other management practices, resulting in greater accumulation of Ca and Mg in the topsoil.

Conclusion
The results of the analysis showed that differences in coffee-based agroforestry management in UB forests affected some of the chemical properties tested.Differences in litter input and microclimate are the factors that cause differences in management.The results of these studies can be used as a basis for determining the best management that can be implemented in UB forests, especially for tree management. 3

Figure 1 .
Figure 1.Study site in UB Forest

Figure 4 .
Figure 4. Soil organic carbon (a) and soil total N (b) under different tree management Note: The data are significantly different based on Fisher's LSD test, which were performed on different letters.Tree management: T1: pine-trimmed, coffee unpruned, T2: pine-untrimmed, coffee pruned, T3: pine-untrimmed, coffee unpruned.

Figure 5 .
Figure 5. Soil total P (a), available P (b), total K (c) and exchangeable K (b) under different tree management Note: The data are significantly different based on Fisher's LSD test, which were performed on different letters.Tree management: T1: pine-trimmed, coffee unpruned, T2: pine-untrimmed, coffee pruned, T3: pine-untrimmed, coffee unpruned.