*3.4. Profitability*

Sole maize had a mean annual investment cost of 670 USD ha−<sup>1</sup> , while that of the sole-longan and the LMG system was 3.7-fold and 3.2-fold higher, respectively. The average maintenance cost of SL and LMG was 300 and 863 USD ha−<sup>1</sup> year−<sup>1</sup> , respectively (Figure 6a). The net profit was related to the cropping system, with an interaction between cropping system and year (*p* < 0.05). The mean net profit of LMG (1018 USD ha−<sup>1</sup> ) was 2.4-fold higher than for SM, while the SL system only achieved a positive profit from year 6 (Table 3). The trend of decreasing net profit of SM across year was partially due to the decreasing selling price of maize over time (presented in the Supplementary Materials Table S2) and lower maize yield in the subsequent compared to the initial years of experiment. From year 2, the net profit from LGM was equal to that from SM, while from year 4 the net profit from LGM was significantly (*p* < 0.05) higher than for SM and SL. In addition, the cumulative profit from LMG was positive from year 2 and higher than that from SM from year 4 (Figure 6a). In contrast, the cumulative profit from SL was still negative in year 7.



SM: sole-crop maize, SL: sole-crop longan, LMG: longan–maize–forage grass, SST: sole-crop son tra, STG: son tra–guinea grass, STM: son tra–mulato grass; na: not applicable since the experiment was established in 2013. Values are means; different letters indicate significant differences (*p* < 0.05).

In the year of establishment, the total input costs were approximately 1772 USD ha−<sup>1</sup> for both STG and STM, but lower (1422 USD ha−<sup>1</sup> ) for SST. In the following years, STG and STM required higher investment than the sole-tree system, mainly deriving from labour costs for forage-grass harvesting (Figure 6b). There was a significant effect (*p* < 0.05) of cropping system on net profit, with the mean net profit in STG (2233 USD ha−<sup>1</sup> ) and STM (2649 USD ha−<sup>1</sup> ) being around 3.2- and 3.7-fold higher, respectively, than in SST (Table 3). The SST system gave a positive net profit from year 3, but the cumulative profit from STG and STM was positive and higher than from SST from year 2 (Figure 6b).

### *3.5. Farmers' Perceptions and Aspirations for Fruit Tree-Based Agroforestry*

#### 3.5.1. Tree and Crop Performance in Agroforestry

Most farmers were fully aware of possible effects of competition for resources (light, water, nutrients) on the performance of tree and crop components within the agroforestry systems (Figure 7). All interviewees in Van Chan noted that growth and productivity of maize in the longan–maize–forage grass system were lower than in sole-maize cultivation. They attributed this to close distance between trees, crops, and grass leading to competition in the agroforestry system. However, non-hosting village groups claimed that longan trees performed better in agroforestry than in sole cultivation since they believed that longan trees utilised the nutrients applied to the maize. However, the experiment-hosting village, the experiment-hosting farmers in Van Chan, and all interviewees in Tuan Giao reported that growth and productivity of trees in both agroforestry systems were lower than when trees were grown separately.

*3.4. Profitability* 

In LMG, the yield of maize grain between grass strips was 24% higher (*p* < 0.05) than in SM and about 62% higher than in upslope and downslope maize in LMG (Figure 5b). Yield of stover was also significantly higher (53–59%) between grass strips than for maize upslope and downslope of grass strips. Overall, the results clearly showed competition between grass, longan, and maize upslope and

Sole maize had a mean annual investment cost of 670 USD ha−1, while that of the sole-longan and the LMG system was 3.7-fold and 3.2-fold higher, respectively. The average maintenance cost of SL and LMG was 300 and 863 USD ha−1 year−1, respectively (Figure 6a). The net profit was related to the cropping system, with an interaction between cropping system and year (*p* < 0.05). The mean net profit of LMG (1018 USD ha−1) was 2.4-fold higher than for SM, while the SL system only achieved a positive profit from year 6 (Table 3). The trend of decreasing net profit of SM across year was partially due to the decreasing selling price of maize over time (presented in the Supplementary Materials Table S2) and lower maize yield in the subsequent compared to the initial years of experiment. From year 2, the net profit from LGM was equal to that from SM, while from year 4 the net profit from LGM was significantly (*p* < 0.05) higher than for SM and SL. In addition, the cumulative profit from

downslope of the grass strips within the LMG system in year 7.

**Figure 6.** Input costs, income, and cumulative profit from: (**a**) the longan–maize–forage grass (LGM) compared with sole maize (SM) and sole longan (SL); (**b**) the son tra–guinea grass (STG) and son tra– mulato grass (STM) compared with sole son tra (SST). **Figure 6.** Input costs, income, and cumulative profit from: (**a**) the longan–maize–forage grass (LGM) compared with sole maize (SM) and sole longan (SL); (**b**) the son tra–guinea grass (STG) and sontra–mulato grass (STM) compared with sole son tra (SST).

**Table 3.** Net profit from the agroforestry systems and the corresponding sole crop/tree. **Net Profit (USD ha−1) Cropping System 2012 2013 2014 2015 2016 2017 2018 Mean (±SE)**  SM 1118a 611a 388a 246b 233b 196b 190b 425.9 (±118.9)b SL −2463c −355b −229b 40b −41b 112b 947ab −284.4 The interviewees also suggested that the agroforestry systems could be optimised through better management of trees and crops (Figure 7). The groups proposed different solutions to improve the efficiency, such as adding more fertilisers to plants suffering from nutrient deficiency in areas where trees, crops, and grass affected each other's nutrient availability, reducing tree density and pruning to reduce shading. In addition, modifying the planting distance between trees and grass was suggested by groups from both sites. The farmers interviewed also suggested less-competitive crops for the

SST na −1422 −290 42 2120 538 3238 704.5

STG na −1772 3297 2853 3069 2381 3570 2232.9

STM na −1772 2661 3018 4067 3147 4773 2648.7

SM: sole-crop maize, SL: sole-crop longan, LMG: longan–maize–forage grass, SST: sole-crop son tra, STG: son tra–guinea grass, STM: son tra–mulato grass; na: not applicable since the experiment was established in 2013. Values are means; different letters indicate significant differences (*p* < 0.05).

In the year of establishment, the total input costs were approximately 1772 USD ha−1 for both STG and STM, but lower (1422 USD ha−1) for SST. In the following years, STG and STM required higher investment than the sole-tree system, mainly deriving from labour costs for forage-grass harvesting (Figure 6b). There was a significant effect (*p* < 0.05) of cropping system on net profit, with

By cropping system *p-*value < 0.001 Cropping system x year *p-*value < 0.001

By cropping system *p*-value = 0.005 Cropping system x year *p*-value = 0.72 (±336.8)c

(±231.6)a

(±632.4)b

(±746.6)a

(±857.1)a

6b).

agroforestry systems, e.g., legume species with biological N-fixation such as soybean and groundnut in LMG in Van Chan (3 of 7 groups), and upland rice or cucumber in STG and STM in Tuan Giao (2 of 7 groups). suggested by groups from both sites. The farmers interviewed also suggested less-competitive crops for the agroforestry systems, e.g., legume species with biological N-fixation such as soybean and groundnut in LMG in Van Chan (3 of 7 groups), and upland rice or cucumber in STG and STM in Tuan Giao (2 of 7 groups).

to reduce shading. In addition, modifying the planting distance between trees and grass was

The interviewees also suggested that the agroforestry systems could be optimised through better management of trees and crops (Figure 7). The groups proposed different solutions to improve the

*Land* **2020**, *9*, x FOR PEER REVIEW 13 of 24

the mean net profit in STG (2233 USD ha−1) and STM (2649 USD ha−1) being around 3.2- and 3.7-fold higher, respectively, than in SST (Table 3). The SST system gave a positive net profit from year 3, but the cumulative profit from STG and STM was positive and higher than from SST from year 2 (Figure

Most farmers were fully aware of possible effects of competition for resources (light, water, nutrients) on the performance of tree and crop components within the agroforestry systems (Figure 7). All interviewees in Van Chan noted that growth and productivity of maize in the longan–maize– forage grass system were lower than in sole-maize cultivation. They attributed this to close distance between trees, crops, and grass leading to competition in the agroforestry system. However, nonhosting village groups claimed that longan trees performed better in agroforestry than in sole cultivation since they believed that longan trees utilised the nutrients applied to the maize. However, the experiment-hosting village, the experiment-hosting farmers in Van Chan, and all interviewees in Tuan Giao reported that growth and productivity of trees in both agroforestry systems were lower

*3.5. Farmers' Perceptions and Aspirations for Fruit Tree-Based Agroforestry* 

3.5.1. Tree and Crop Performance in Agroforestry

than when trees were grown separately.

**Figure 7.** Farmers' perception of the performance of trees and crops in the agroforestry experiments in Van Chan and Tuan Giao compared with that of sole crops/trees. Open-ended questions were used in group interviews with non-hosting and experiment-hosting villages; experiment-hosting farmers were interviewed individually.
