Water supply on the island of Lombok overall is stressed, including the supply for the capital city of Mataram. The highlands surrounding the capital are forested and mostly undeveloped, while the lowlands are highly cultivated. The southern part of the island is fertile but drier, especially toward the southern coastline. Agriculture is main economic sector in the island of Lombok.
The province is one of the Indonesia’s main crop producers, particularly rice. Annually, West Nusa Tenggara produces more than 40 thousand tons of rice, with most of the yield produced on Lombok Island. Yet around 16 percent of all paddy fields on the island are rainfed and would be greatly affected in times of longer droughts and diminishing precipitation. The rest are naturally affected by water scarcity as a direct result of rising temperatures and the destruction of forests protecting the watershed areas.
The frequency of large droughts in the country increased over the last 40 years compared to the previous decade - from once in three or four years to once in two or three years (Boer and Subbiah, 2005). Similar observations have been made for floods. Based on historical data from 1989-2008, rice crop failures due to drought increased significantly during El Nino years, particularly in West Lombok (based on data from Directorate of Plant Protection, Ministry of Agriculture).
The National Action PLan for climate change adaptation (RAN-API, 2013) for Indonesia identifies agriculture, estate crops, forestry, fisheries and coastal, health and fresh water as sectors of concern in Lombok and pilot vulnerability assessments have identifies shortages of clean water, tidal floods and abrasion as clusters of action. The RAN-API concerns in their National Action Plan for food security is decrease in agricultural and fisheries production due to climate change; identification of new growth areas of food production in low climate risk areas; development of food security systems for farmers, fishermen and community.
Defining the question
As Central Lombok is the region where agriculture is the main livelihood this analysis will focus on this region. Feedback for participants at SEI led training in November 2013 in Mataram requested further information on conducting a local climate and vulnerability analysis and this module aims to address this gap. The method used follows weADAPT 's guide to using climate information for adaptation planning.
Lombok utilizes about 80% of its water for crops. Rice production in NusaTenggara Barat (NTB) has increased in both area and yield over the past 50 years. Similarly, the population in the province has also increased. These two factors, rice area increase, and human population increase, are major contributors to the current water stresses that are found in Nusa Tenggara Barat.
According to the data report from Central Bureau of Statistics (CBS) of Nusa Tenggara Barat province, for the last nine years (2004- 2012), productivity of crop, decreased in both the land harvest area (ha) and production quantities (ton). This is thought to be due to changes and shifting of seasons. One of the main indicators of shifting seasons is the irregular rainfall pattern. This phenomena makes the farmers in NTB, especially Lombok, experience difficulties in determining cropping patterns (Irawan et al., 2013).
In terms of year to year variability and longer term changes in precipitation, planting failure will occur if there is a lack of water availability in early to mid November. Growing failure will occur if there is lack of water in mid-end December and harvest failure will occur when rice stems collapse due to heavy rain and strong wind or floods.
The question to be addressed is therefore, how are changes in climate and in particular rainfall pattern likely to affect rice production in Central Lombok?
Historical climate data from the weADAPT-CIP interface
Ampenan/Salepar is the nearest station to Central Lombok. It is situated close to the city of Mataram. It is in fact the only station on the CIP network for Lombok.The total rainfall from this station is on aver 1100mm per year. The historical climate monthly averages for Ampenan/Selapar (see below) indicate a wet season from November to March, followed by a dry period from June to September. April, May and October are intermediate in terms of total precipitation with approximately 60-80mm falling in those months, while November and February are the wettest months. Average monthly temperatures show a narrow range throughout the year from approximately 29.5oC in July to maximum of over 31oC in April and October.
Historical monthly averages in Tmax, Tmin and rainfall 1977-2007
Observed temperature changes
In terms of temperature trends over the observed period, the concern of increases in daytime maimum and nightime minimum is confirmed by the daily maximum temperature graph. Daily maximum temperatures have increased particularly since 1994 with daily maximum often exceeding 32oC, which was rare before this date. The minimum temperature however seems to have remained stable.
Average maximum temperature 1977 -2007
If we look at the rainfall data, it is hard to discern trends in monthly precipitation and the number of rainy days greater than 10mm. It is also not possible to tell in the historic record if there has been a shift in the start of the monsoon season. Missing data in the record does not allow for analysis of how the precipitation has changed during the growing season. The Ministry of the Environment (2010) noted that January rainfall has become lower since the 1990s. This can be noted in the rainfall data from 2000 to 2007 in terms of both monthly totals and heavy rain days (>10mm).
Total monthly rainfall 1977-2007
Total monthly rain days with >10mm 1997-2007
Future climate projections - precipitation changes
The figures below show the range of projected future changes (in total monthly rainfall ) across 7 statistically downscaled CMIP5 GCMs for Mataram for two scenarios SRES A2 and B1 for mid century. The values plotted are not absolute rainfall totals but rather “anomalies” (or changes) in the model simulations from the average monthly rainfall in a historical period (1977 to 2000) to the selected future period (2045 -64). For example, in June the lowest value (-30mm) shows the change between the historical period and the future period of the model simulation, corresponding to an average drying. The red bars show the difference between the zero change level and the 10th percentile values while the blue bars show the difference between the zero change level and the 90th percentile. When there are is no blue bar (e.g. August) or no red bar (eg April) then at least 90% of the models agree on the sign of the change (e.g. drying in August). These are model to model anomalies; there is no information provided about the reliability of each model used and whether they accurately represent the observed climate at the station.
In most months there is a range of change somewhere between plus 20mm and minus 20 mm. Despite the recent observations of changes in January the projections show a bias to higher rainfall in January. The biggest ranges in projections are in months of January, May, June November and December for A2. B1 has a stronger drying trend across all months. August again shows a consistent drying trend and April a wetter trend across the models used. This could possibly suggest a contraction of the rainy season, with less rain falling in the intermediate months of May and October. If rainfall does decrease in the growing period November to March which looks possible in the B1 scenario there will be even more pressure on the coping strategies of central Lombok to manage their water. Responding to interannual variability is therefore likely to remain a priority for farmers in the region.
Total monthly rainfall for SRES A2 Scenario for 2046 to 2065
Total monthly rainfall for SRES B1 scenario for the period 2046 to 2065 (downscaled CMIP3). Anomalies are relative to the historical period 1979-2000. The solid bars represent the range between the middlle 80% of projected changes. Grey lines show the projected change for each of several models
A scenario for monthly precipitation for the end of the century indicates a more worrying picture. The drying trend continues for the B1 scenario and there is consistent drying for the months of December and January, when vigourous growth should occur.
Total monthly rainfall under the SRES A2 scenario for 2081-2100
Total monthly rainfall under the SRES B1 scenario for 2081-2100
Discussion and lessons learnt
These observed trends in precipitation are of concern for dryland rice growing in central and southern Lombok. The gogorancah (Klock and Sjah, 2011) system for dryland production is a system where dry seeding takes place after two or three rainfall events. If rain does not start consistently in November and continue December, there will be consequences for yields. Increasing maximum temperatures continue will also have implication for grain filling and final yields.
The model results in terms of warming and drying reflect an earlier study that looked at assessing the risks of climate change on Indonesia rice production which suggests that under future climate projections, finding a significant 30-day delay in the onset of monsoon season and a substantial decrease in precipitation later in the dry season (Naylor et al., 2007), which when combined with temperature increases of up to 4°C (for every 1°C increase in minimum temperature, rice yields can decrease by 10%; Peng et al., 2004), could lead to massive drops in rice production.
In terms of the climate data available through weADAPT and the Climate Information Portal, the Indonesian data requires contains lots of missing data; it requires updating to more recent climate change scenarios and there are missing scenarios for changes in temperature and number of wet days, all would add to this analysis. Dry events are particularly associated with El Nino and better anticipation of these events would allow for improved planning for production. More detailed understanding of the planting regime and water requirements during the rice development would also improve the analysis.
Measures to improve water use and soil retension of water are likely to be very useful, and a focus on adapting to inter-annual variability in precipitation will help with the changes that seem likely in the scenarios.
Robinson,B., McClymont ,D., Abawi, Y. and D. Rattray. Irrigation Practice and Infrastructure Design in the Variable Monsoonal Climate of Lombok (Indonesia). Queensland Climate Change Centre of Excellence, Natural Resources and Water, Queensland.
Irawan, M. et al. (2013) Irrigation water requirement system based on artificial neural networks and profit optimisation for planting time decision making of crops in Lombok island. Journal of Theoretical and Applied Information Technology. 31 December 2013. Vol. 58 No.3. ISSN: 1992-8645 www.jatit.org.
Klock, J., Sjah, T. (2011) Farmer Water Management Strategies for Dry Season Water Shortages in Central Lombok, Indonesia. Natural Resources, 2011, 2, 114-124. doi:10.4236/nr.2011.22016 (http://www.scirp.org/journal/nr). Copyright © 2011 SciRes. NR.
World Food Programme (2012) Indonesia: Strengthening Community Resilience To Climate Change In Lombok By Mr. Teuku Yunansyah — 14 June 2012
WWF (2013) Poster: http://awsassets.wwf.or.id/downloads/adaptasilombok.pdf