FARMING INPUTS AND WATER STRESS IMPACT ON GROWTH AND YIELD OF GARLIC (Allium sativum L.) IN THE RAINFED HIGHLANDS OF EASTERN LOMBOK, INDONESIA Baiq Nurul Hidayah, Mardiana, Sylvia Kusumaputri Utami, Nani Herawati, Titin Sugianti, Darwis and Muhammad Faesal Matenggomena Indonesian Agency for Agricultural Research and Development (IAARD)-The Institute for Assessment of Agricultural Technology (Balai Pengkajian Teknologi Pertanian) West Nusa Tenggara Province, Indonesia

Baiq Nurul Hidayah, Mardiana, Sylvia Kusumaputri Utami, Nani Herawati, Titin Sugianti, Darwis and Muhammad Faesal Matenggomena Indonesian Agency for Agricultural Research and Development (IAARD) The Institute for Assessment of Agricultural Technology (Balai Pengkajian Teknologi Pertanian) West Nusa Tenggara Province, Indonesia. ...................................................................................................................... Manuscript Info Abstract ......................... ........................................................................ Manuscript History Received: 15 May 2020 Final Accepted: 20 June 2020 Published: July 2020

Garlic (Allium sativum L.) is a horticultural commodity of high economic value in Indonesia. West Nusa Tenggara (WNT) Province is one of the largest garlic producing regions in Indonesia with 45% to the national harvest area. Garlic is mostly grown in irrigated highlands; however domestic production is very limited and only supplies at most 5% of domestic demand. Currently, garlic farming is facing the problem of decreasing soil fertility, the use of high inputs of inorganic fertilizers, and excessive chemical pesticides. Therefore, it is beneficial to investigate garlic cultivation input packages that may reduce inputs and so lower production costs in rainfed highlands which are prospective areas for garlic. The input-saving packages investigated were reduced use of inorganic fertilizers and chemical pesticides; up to 50% from farmers' practices. The four applied input packages were P1 (farmers' existing practices) and reduced and modified input packages P2, P3 and P4. The experiment was conducted in the Sembalun Highlands (1200 m above see level) of Eastern Lombok, WNT Province during the growing season of 2018 on 0.4 ha of rainfed land using a completely randomized block design. Parameters measured were agronomic performance and productivity. Results showed that there were no significant differences in the agronomic growth and productivity of garlic from the four input packages applied. However, the experiment had sub-optimal yield due to the early onset of the dry season a because of climate change.
The Sembalun Highlands in Eastern Lombok, West Nusa Tenggara Province are one of the garlic centers of Indonesia. From 1997 until 2017 garlic cultivation in the region declined as it is uncompetitive with the cheaper imported garlic. In the last few years, the cultivation of garlic was revived by the Indonesian government. One of the strategies to increase domestic production is to expand the growing areas into wet season production in rainfed highlands. The objectives of this research were to investigate agronomic performances and productivity of garlic crops (local Sangga Sembalun cultivar) based on different farming input packages in the rainfed highlands of Eastern Lombok, Indonesia.

Methods
A field experiment was conducted in the Sembalun Higlands (1200 m asl) in Eastern Lombok of Indonesia from February to June 2018 to investigate cultivation input packages on agronomic performance and yield of a local variety of garlic. The experimental design was a completely randomized block of four blocks. Each block consisted of four treatments: 1) Farmers' usual cultivation input package (P1); 2) Cultivation Input Package I (P2); 3) Cultivation Input Package II (P3); and 4) Cultivation Input Package III (P4). Therefore, there were sixteen plots in total. The experimental was 0.4 ha comprising four replication blocks of 0.1 ha each. The details of the cultivation input package are shown in Table 1. The experimental randomisation is shown in Table 2. Spraying pesticides based on rotation strategies Table 2:-The experimental design and plot randomization in the field during the growing season of 2018 BLOCK  I  II  III  IV  P3  P1  P4  P2  850   P1  P2  P3  P4  P4  P3  P2  P1  P2 P4 P1 P3

Statistical Analysis
Data was analysed with SPSS version 12.0 using one-way ANOVA followed by least significant different (LSD) test at the 95% confident level (p<0.05).

General condition of the experimental site
The experimental site was a farmers' field with horticulture commodity cropping patterns; the sandy sand soil had a pH of 5.5. The location is a rainy season horticultural area in East Lombok Regency, categorised as a rainfed highland. The crop's water needs relied on rainfall. Soil moisture at sowing was suitable for planting garlic, but as the crop matured and approached harvest there was no rain resulting in water stress. The map of the experimental site is shown in Figure 1. The soil type of the experimental site based on soil analysis in the Institute for Assessment of Agricultural Technology (IAAT) West Nusa Tenggara laboratory was sandy sand where sand was more dominant than silt and clay. The pH below 6 was ideal site for treatment of dolomite for increasing pH in order to make the soil more productive (Rochayati et al., 1986).

Vegetative growth
The vegetative growth of the garlic treatments was observed at 4, 6 and 8 weeks after sowing (WAS). Parameters measured were plant height and leaf width. Plant height and leaf width for each treatment are presented in Figures 2 and 3. Figure 2 shows the height of the treatments at 4, 6 and 8 WAS did not show any statistically significant difference. This shows that although P4 supplied the minimum input plant height was not significantly different from the higher input treatments. Figure 3 shows that there was also no significant difference in leaf width between treatments. Thus, the minimum cultivation input package (P4) produced the same leaf width as the other input package treatments.

Garlic yield
Garlic yield of four cultivation input packages during the field experiment in 2018 showed suboptimal yield due to the environmental factor of water availability. Rainfall was very low from the beginning of April 2018 until harvest. Data on garlic productivity, both fresh and dry weight (1 month after drying) is shown in Figure 5. Figure 5 shows that garlic fresh weight was not significantly different for each cultivation input package. Likewise, dry weight after sun drying for one month showed no significantly difference. Although this yield was not optimal due to the early onset of the dry season, the yield of both fresh and dry weight shows that P4 produced garlic that was not significantly different from the treatments with higher cultivation inputs. The water deficit happened when the garlic treatments had started bulb formation and required plentiful water. Water is very important to obtain maximum yield especially during bulb formation (Sadaria et al., 1997). Bulbs tend to have reduced size if water stress occurs during bulb formation (Fabeiro Cortes et al., 2003). In addition, Nam et al. (2007) stated that water stress influences both agronomic growth and bulb size.
The environmental constraints which occurred during the field experiment in growing season of 2018 were very low rainfall during the cropping phase that requires lots of water for bulb formation and development. The lower rainfall and fewer rainy days that occurred from early April 2018 when crop growing conditions entered a generative phase that required plentiful water caused the suboptimal growth of the crops and development of bulbs was disrupted.
When we compared garlic yield based on the farmers' existing practices (P1) with farmers' yield during a normal season, the garlic yield was reduced by almost 40-50% due to water stress.

Environmental constraints
This 2018 field experiment was affected by a lack of rain from the beginning of April when the crops were 45 days old and begining bulb formation. Treatments which were planted from the 17-18 February received sufficient rainfall during early vegetative growth, but from early April rainfall in the Sembalun highlands was far below average which meant the crops were water stressed during the period when they needed water most. Rainfall data for the Sembalun highlands during the growth of the field experiment can be seen in Figures 6 and 7.   The maximum temperature in April 2018 reached 27.6 o C while in the same month of the previous year it was hald a degree lower at 27.1 o C. We also felt this temperature increase when we were in the field due to the wind accompanied by heat waves which also affected the speed of the crops were dry out as an additional result after the absence of rainfall. However, there were no data that indicated the level of wind speed accompanied by heat waves due to the unavailability of tools in the field to measure the data. Anonymous (1998) stated that garlic crops adaptive in temperature between 15-20 o C. If the temperature is below 15 o C, the crop growth is disrupted and slow, while if the temperature is above 27 o C bulb formation is also disrupted.   855 The plant height of all garlic treatments at 4, 6 and 8 WAS did not show significant differences. Furthermore, leaf width also was not significanlty difference between treatments.Treatment yield at harvest (fresh weight) was not significantly different for each cultivation input package. Likewise, dry weight of bulbs after sun drying for one month showed no significantly difference between treatments. Although these yields were suboptimal due to the early onset of the dry season, both fresh and dry weight of the lowest cultivation input package (P4) was not significantly different from the treatments with higher cultivation inputs.