Farmers Willingness to Accept (Wta) for Submergence Rice Varieties at Flash Flood and Flood Prone Affected Rice Area

Farmers are rational in decision making process with respect to any introduced agricultural technology. Farmers may consider the economic sacrifices in term of additional cost and potential benefit or additional income before they accept and adopt the introduced technology. This study aimed to analyze farmer's criteria and determine explanatory variables affecting farmer's decision to accept or to adopt submergence tolerant (Sub-1) rice varieties at flash flood and flood prone affected rice area. The study was conducted in Indramayu District, West Java, and Kayu Agung District, South Sumatra. Contingent Valuation Method (CVM) that derived Willingness to Accept (WTA) approach was exercised to analyze explanatory variables that influence farmers' willingness to accept introduced rice varieties. The results showed that the economic cost of flooding that damaged rice was about US$7.63 million in Kayu Agung and US$11.25 million in Indramayu in every wet season planting. Farmer's criteria used in submergence tolerant varietal evaluation varied and location specific in nature. Most of explanatory variables used in the model were significantly influenced farmers' WTA for submergence tolerant rice varieties such as: (1) availability of seed, (2) submergence tolerant for more than 14 days, (3) high yield, (4) proffer rice taste, (5) households' income during normal year, (6) area planted during normal year, and (7) age of farm household head. The only indicator that did not significantly influence the farmers' WTA for the Sub-1 rice varieties was farm household income during the flood year cropping.


ABSTRACT
Farmers are rational in decision making process with respect to any introduced agricultural technology. Farmers may consider the economic sacrifices in term of additional cost and potential benefit or additional income before they accept and adopt the introduced technology. This study aimed to analyze farmer's criteria and determine explanatory variables affecting farmer's decision to accept or to adopt submergence tolerant (Sub-1) rice varieties at flash flood and flood prone affected rice area. The study was conducted in Indramayu District, West Java, and Kayu Agung District, South Sumatra. Contingent Valuation Method (CVM) that derived Willingness to Accept (WTA) approach was exercised to analyze explanatory variables that influence farmers' willingness to accept introduced rice varieties. The results showed that the economic cost of flooding that damaged rice was about US$7.63 million in Kayu Agung and US$11.25 million in Indramayu in every wet season planting. Farmer's criteria used in submergence tolerant varietal evaluation varied and location specific in nature. Most of explanatory variables used in the model were significantly influenced farmers' WTA for submergence tolerant rice varieties such as: (1) availability of seed, (2) submergence tolerant for more than 14 days, (3) high yield, (4) proffer rice taste, (5) households' income during normal year, (6) area planted during normal year, and (7) age of farm household head. The only indicator that did not significantly influence the farmers' WTA for the Sub-1 rice varieties was farm household income during the flood year cropping.

INTRODUCTION
The lowland rice-producing areas of the world may be classified first by water regime, namely those with a reliable and controlled external supply of water and a drainage system, the irrigated rice lands, and those depending solely on rainfall and runoff, the rainfed lowlands. The later are diverse and range from drought-prone lands to those subjected to flooding in excess of several meters during the growing season. Rice production in these areas has not really bene-fited from the technological advances that were so successful for irrigated rice, yet hundreds of millions of Asia's poorest people depend upon the rainfed lowlands for their livelihood (Zeigler and Puckridge 1995).
Flooding is a major stress constraint to rice production, especially in rainfed lowland areas of the tropics. The flooding of the major river basins and deltas of Asia has provided the sustenance for the rice production that has been a prominent feature of the region for millennia. However, flooding is also a cause of yield fluctuations because of erratic rainfall patterns and poor drainage of many rice fields. This results in excess water in these fields for varying depths and durations . For convenience, this flooding can be classified into four types depending on the plant traits and varietal types adapted to the conditions, namely: (1) flooding during germination (anaerobic germination, a problem when direct seeding is practiced and heavy rains result in submergence before germination; (2) flash flood (submergence), plants are completely submerged for up to two weeks thus submergence tolerance is required for this condition; (3) stagnant flooding (medium deep or semi-deep), flooding occurs for a longer duration, more than two weeks and often several months, at depths up to 50 cm thus it requires varieties tolerant to stagnant flooding conditions; and (4) deeper stagnant flooding (deepwater or floating rice), in which water depth increases throughout the season to depths of above 50 cm and often a meter or more. Varieties with tall plant height or rapid internodes elongation are required .
Submergence is a recurring problem in the riceproducing rainfed lowlands of South and South-East Asia. Developing rice cultivars with tolerance submergence and with agronomic and quality traits acceptable to farmers is a feasible approach to address this problem. In areas where high-yielding but submergence-intolerant rice varieties have been cultivated, farmers suffer from crop losses caused by periodic flash floods during the monsoon season (Septiningsih et al. 2009). Recently, the extent of submergence stress has increased due to extreme weather events such as unexpected heavy rains that have inundated wider areas across many regions in Asia such as Indonesia. More sustainable solutions are needed to overcome this problem. One of the most promising solutions is to develop high-yielding varieties that are submergence tolerant and likely to be adopted by the farmers in the target regions.
Through collaborative research led by the International Rice Research Institute (IRRI) with National Agricultural Research (NARs), a flood-tolerant local rice variety was investigated to isolate the gene responsible for flood resistance. Using a technique known as marker-assisted backcrossing, scientists transferred the water tolerant trait of interest into commercially valuable local rice varieties without losing useful characteristics which make them popular with farmers. Identification of the genetic code of the SUB1 gene controlling submergence tolerance enabled breeding of new, high-yielding Sub-1 mega-varieties, with popular characteristics such as high yield, good grain quality and local pest and disease resistance (IRRI 2011). Fortunately, local rice landraces cherished by farmers include accessions adapted to extremes in water availability, including tolerance to progressive flooding or rapid submergence that can be the source of genetic variation to be used to improve the adaptability of rice to extreme environmental stresses such as submergence (Serres et al. 2010).
Indonesia is one of the country participants based on characteristics and targeted output of the project, Until present condition, not much effort has been implemented to fully understand farmers' production constraints and challenges both institutional and technological factors. Thus, this may hinder the adoption of the varieties to be disseminated at each area as well as other impact areas. Farmers may take into consideration the balance between added cost and potential benefit and return before they will to accept and adopt an introduced technology (Manikmas et al. 2009).
Willingness to Accept (WTA) approach is a measure of the economic scarifies in terms of income or other goods a person is willing and able to forgo gain or maintain a resource, good, or service. Net WTA, which is the difference between WTA and actual action for introduced technology and its impact to the ecological services, is usually used for cost-benefit analysis. Whether WTA is actually collected as cash is largely irrelevant from the standpoint of economic efficiency. While it may be important for political reasons to transfer a portion of the user's WTA to actual cash flow, any financial returns are just a transfer of benefits from user to recipient. The total economic value received by society does not change, only the distribution of the economic value among members of society (Willing 1976).
Technology adoption is governed by farmer perceptions regarding advantages and disadvantages of new technology compared with the existing technology. Yet farmer perceptions are not adequately documented and analyzed in adoption studies. In this paper, farmer perceptions regarding traditional and improved rice varieties in a rainfed village in Orrisa are used as a basis for explaining the low adoption of improved varieties and the high degree of varietal diversity. The results indicate that quality characteristics loom very large in farmers' choice of rice varieties (Kshirsagar et al. 2002). Quality characteristics such as seed purity, grain and milled rice quality, yield potential of submergence tolerant rice varieties depend upon its genetic background.
This study aimed to determine the economic cost and farmer's criteria in evaluating introduced varieties and analyze the determining explanatory variables that potentially affecting farmer's decision to accept or to adopt submergence tolerant rice varieties at flash flood and flood prone affected rice area.

Locations and Respondents
The sudy was conducted in Indramayu District, West Java, and Kayu Agung District, South Sumatra. Indramayu District was selected as a key site representing flash flooding areas along the North Coast of Java. Meanwhile, Kayu Agung District was selected to represent flood prone area as a satellite site. Direct and face-to-face individual interview was conducted in both districts. Participatory varietal selection (PVS) was exercised to derive farmers' WTA for newly introduced submergence tolerance rice varieties.
Stratified random sampling was applied to cluster the farm household samples. The cluster was based on few indicators such as income class, size of land holding and level of flooding. The number of respondents was 100 farmers in Indramayu and 60 farmers in Kayu Agung for a total of 160 respondents that

Data Collection
Baseline surveys was exercised to collect data and relevant information at farm household level at targeted study areas. This type of survey aimed to establish initial or ex-ante condition against which the effects of a finished project can be compared or ex-post condition. These study efforts are used to provide a quick snap shot of the sector and to understand processes where a project would be involved in and serve as benchmark against which the impact of a project will be measured (NB Ideas 2007).
In both cases, some of the data required to measure the impacts of any new technology innovation at aggregate level can be directly observed, and other data must be estimated indirectly from other evidence. Choosing and using data are most important skill required in baseline study. The difference between successful and unsuccessful of ex-ante assessment typically rest on the judgment of the researcher in collecting and interpreting their data (Master et al. 1996).
Well structured questionnaire was used to record all possible factors affecting WTA for introduced Sub-1 rice varieties, included: 1) availability of seed, (2) submergence tolerance at most 14 days, (3) high yield, (4) suitable rice taste, (5) household income at normal year, (6) on-farm income at flood year, (7) area affected by flood, (8) paddy price, (9) age of household head, (10) education of household head, and (11) number of household members. Economic cost of flooding and farmer's criteria in evaluating new rice varieties were other considerably important factors in evaluating the adoption opportunity of newly introduced rice varieties.

Approach
Contingent Valuation Method (CVM) has been widely used in agricultural research and development. In this method, farmers can be directly interviewed individually by asking how they would be willing to pay or to accept for the improvements of their practices or the environmental impact which agriculture technology brings (Yoshida et al. 1997;Yoshida 1999). CVM uses surveys in which people are asked how they are willing to accept (WTA) for a certain level of agricultural technological change through introduction or improvement of the existing technology. The basic notion underlying CVM is that a realistic but hypothetical condition for targeted group to accept a certain technological change can be described to an individual. Key features of the hypothetical change include: (1) description of the resource being valued; (2) means of acceptance, such as an increase in yield and quality; and (3) the value of elicitation. The means of acceptance level must be realistic and emotionally neutral for the respondents. To improve realism, the means of acceptance must be appropriate for the technology.
In principal, application of CVM is based on potential behavior, which is not based on actual behavior (Munasinghe 1992). The application of CVM is basically assessing the farmers' technological preferences with respect to the observable benefit through investigation of farmer's WTA. CVM has been shown to be reliable, especially for estimating user values. CVM is generally consistent with axioms of revealed choice at least for familiar goods and services. CVM typically has been shown to compare very favorably with other non-market resource valuation techniques (Cummings et al. 1986;Mitchell and Carson 1989).
CVM is the most widely exercised approach for determining the value of non-market goods and agricultural technological impact to the end users. CVM has been shown to produce benefit estimates close to the true benefit and is also reliable in retesting at least for user values (Loomis 1987). Introduction of agricultural technology often produces both costs and benefits that may not be valued totally. In addition, some government agricultural programs not yet initiated may have unknown cost that cannot be summarized from the program activity. This approach is also intensively exercised to evaluate the willingness of community living along the Citarum watershed area to the programs offered by government to conserve the environment such as reforestation, alley cropping, terracing, water reservoir and household well, irrigation and drainage system, and agroforestry (Manikmas and Fahmudin 2003). Other study also applied CVM to evaluate the WTA of farmers with respect to newly introduced food crop varieties such as red rice, quality protein maize (QPM), pulut maize and sweet potato. In aggregate, similar variables exercised in this study were able to explain 70-90% of farmers WTA to those introduced varieties (Adnyana et al. 2007).

Contingent Valuation Model
Assumes that unobserved WTA could be expressed as: Where ε I is independently and identically distributed (0,δ 2 ), and the X i 's are the explanatory variables. In this study the mathematical equations of farmers' WTA for flash flood affected rice areas in Indramayu are as follow:

Data Analysis
Ordinary least square (OLS) was used to estimate the parameters for each respondent's WTA. Since OLS has given quite efficient parameter estimate in term of coefficient of determination and magnitude of the estimates, no other technique was exercised. Responsiveness of respondents to the change of explanatory variables was also computed in term of elasticity value since the model used is double log equation.
Four WTA equations were estimated with precondition all sign of parameter estimates are positive except area planted during the flood year. In other words, all explanatory variables included in the model positively influence the variability of farmers' WTA for the introduced submergence tolerant rice varieties. The explanatory variables included in the model were based on the continuous estimation process until the most suitable variables were fit to the model.

RESULTS AND DISCUSSION
Results of the study related to economic cost of flooding and farmer's criteria in evaluating new varieties are presented in Tables 1 and 2. Meanwhile, all estimated parameters of explanatory variables to explain WTA for introduced Sub-1 rice varieties included: (1) availability of seed, (2) submergence tolerant at most 14 days, (3) high yield, (4) suitable rice taste, (5) household income at normal year, (6) on-farm income at flood year, (7) area affected by flood, (8) paddy price, (9) age of household head, (10) education of household head, and (11) number of household members. The WTA model was estimated to explain factors or explanatory variables that influenced farmers in decision process with respect to the introduced submergence tolerant rice varieties. The outputs of WTA analysis are presented in Tables  3 and 4.

Economic Cost of Flooding
Economic cost of flooding that damaged rice areas at aggregate level was counted based on average price of paddy during the baseline study was conducted. The on-farm price of paddy in Kayu Agung was about US$210.2/t, while price of paddy in Indramayu was accounted at US$235.6/t. On the other hand, price of paddy at national level was about US$226.9/t. Based on these paddy prices and average production loss, therefore the total economic cost of flooding was accounted for about US$7.63 million in Kayu Agung and US$34.02 million at provincial level of South Sumatra. On the other hand, cost of flooding in Indramayu was about US$11.25 million and US$38.54 million at provincial level of West Java. Meanwhile, nationally economic cost of flooding was estimated at least US$288.60 million in average per year within last 12 years (Table 1).
In Viet Nam for example, every year, resource-poor rice farmers in the Mekong Delta experience varying types of flooding causing yield loss of about 20-100%. Since rice is the main source of livelihood in the rural areas, frequent floods have contributed to increased poverty (Ngoc Chi et al. 2011b). To avert More precisely, impact of flash flood to rice production in flood year in Mekong Delta, Viet Nam, was relatively lower than those in normal year of about 15%. The rice yield in the flood year also reduced from 4 to 3.73 t ha -1 . The gross household income in the flood year was relatively lower than those in the normal year (about 18%). Of which, the reduction in rice income was 19%. This caused reduction in the net income from rice production in the flood year (Ngoc Chi et al. 2011a).
In addition, submergence caused by typhoons and floods is one of the major reasons for production losses. Because of the complexity of these ecosystems, the breeding framework necessitates adequate feedback and a more in-depth understanding of the ecological and socioeconomic conditions in these flood-prone areas (Manzanilla et al. 2011;Mariano et al. 2012). Within this purview, this study validated the performance in farmers' fields of lines with the SUB1 gene that confers tolerance of submergence for up to two weeks. The evaluation was conducted through participatory approaches to gain understanding of the risks as well as farmers' preferences for these varieties (Manzanilla et al. 2011).

Farmers' Criteria in Evaluating New Varieties
The farmers' perceptions with respect to the criteria they use to select the existing varieties at flood affected rice areas in Indonesia were grouped into five priorities (rank) such as most important (rank 1), important (rank 2), quite important (rank 3), less important (rank 4) and not important (rank 5  Table 2). The benefits of developing and releasing submergence salinity-tolerant and phosphorous-deficiencytolerant rice in Bangladesh, India, Indonesia and the Philippines are estimated for marker-assisted breeding (MAB) and for conventional breeding (CB) using economic surplus analysis. Marker-assisted breeding is estimated to save at least 3-6 years in the breeding cycle and result in incremental economic benefits over 25 years in the range of US$50-900 million compared to CB, depending on the country, stress and time lags. Saline and phosphorus-deficient soils are difficult problems to address through CB because of undesirable traits that accompany desirable ones during the breeding process. Marker-assisted breeding enabled by advances in genomics and molecular mapping is more precise and time-saving. Costs are estimated at US$3.4 million for MAB and US$2.5 million for CB, and hence the additional net benefits of MAB in rice far exceed those for CB (Alpuerto et al. 2009).
Study conducted in the Philippines showed that varying conditions of submergence can influence farmers' criteria and preferences for rice cultivars. Depending on the timing of flood with respect to growth stage, shorter duration and shallow flashfloods result in less than 10% production losses while deeper and stagnant water with two weeks duration and >100 cm depth cause damage ranging from 40% to 77%. Major findings of PVS trials and preference analysis indicated that farmers prefer rice cultivars that are tolerant to submergence, have early to medium maturity relative to their commonly grown varieties, and are resistant to pests, diseases and lodging, among other traits. To enhance adoption, male and female farmers should be involved in the evaluation process. The results of this study can contribute to enhancing breeding programs to develop appropriate varieties that reduce production losses, improve income and ultimately reduce poverty incidence in submergence-prone areas (Manzanilla et al. 2011).

Indramayu District
The overall performance of estimated farmers' WTA for Sub-1 rice varieties in Indramayu during the normal year was statistically acceptable. This was shown by the coefficient of determination (R 2 ) that accounted for about 0.819 and 0.841 during the normal year and flood year, respectively. The accuracy of estimated WTA model was also quite high with coefficient of variation 0.145 for normal year and 0.215 for flood year. The estimated parameters showed that most of explanatory variables used in the model were significantly influenced the farmers WTA such as: (1) availability of seed, (2) submergence tolerant, (3) high yield, (4) rice taste, (5) households' income during normal year, (6) area planted during normal year, and (7) age of household head. The only indicator that did not significantly influence the farmers' WTA for Sub-1 rice varieties was household income during the normal year.
Rice planted area during flood year in Indramayu negatively influenced the farmers' decision in determining their WTA. In other words, if household income increased during the normal year then farmers tend to reduce their WTA for the submergence tolerant rice varieties. This condition was because the introduced Sub-1 rice varieties were not well tested during the flood year or farmers are waiting for the real on-farm performance of these varieties. The estimated parameters of explanatory variable are presented in Table 3.
Availability of seed and the submergence tolerance of introduced Sub-1 rice can be improved at least by 10% then the farmers' WTA will increase at least by 0.438% and 0.894%, consecutively. In addition, the 10% improvement of rice yield and rice taste will also improve the farmers' WTA at least by 0.258% and 0.163%, respectively. In general, if all explanatory variables used in the WTA model improved by 10% each, then farmers' WTA will also increase at least by 2.426% during the normal year wet cropping season. This condition indicates that farmer's response that expressed in WTA is quite slow since farmer may wait for better varieties field performance in relation to availability of seed, submergence tolerance level on introduced varieties, yield, rice taste and on-farm price of harvested paddy during the normal year. Similar to WTA model used to estimate parameter of variables that influence farmers WTA during the normal year, the WTA model used to explain the farmers WTA during the flood year was also statistically valid and significant. About 84.06% of farmers' WTA variability was able to be explained by the model with coefficient of determination (R 2 ) 0.8406 and coefficient of variation (CV) 0.2148. This indicated that the accuracy of estimated parameters was quite high. In other words, the WTA model used in this study was able to represent the farmer's response to the introduced submergence rice varieties planted during flood year.
More factors influenced farmers' WTA during flood year compared with normal year since more problems were encountered by farmers during the flood year. Factors that significantly affected the farmers WTA for submergence tolerant Sub-1 rice varieties during flood year were (1) availability of seed, (2) submergence tolerant up to 14 days, (3) high yield, (4) suitable taste, (5) price of paddy, (6) level of education of household head and (7) number of household member. However, on-farm crop income during flood year, planted area, and age of household head did not significantly affect farmers' WTA.
Four important farming indicators such as seed availability, submergence tolerant up to 14 days, high Nevertheless, level of education and number of household members also affected farmers' WTA for introduced Sub-1 rice varieties with estimated parameter of about 0.0297 and 0.0321 during the flood year respectively. In addition, an overall increase of each indicator or explanatory variable by 10% will increase farmers' WTA during flood year by 7.945%. This figure was almost four times higher compared with farmers' WTA during normal year (Table 3). This indicates that farmers are more responsive to accept the introduced Sub-1 rice varieties during the flood year as one of the coping mechanisms for the crop damage due to flash flood during wet season cropping.
The discussion about farmers WTA for Sub-1 rice varieties in Indramayu conclude that farmer's response to these introduced varieties was relatively higher during the flood year compared with normal year. This indicates that more problems and challenges are faced by farmers during flood year compared with normal year. This high response also indicates that farmers are trying to find out more coping mechanisms during flood year.
To compare, outputs of simlar study conducted in Viet Nam showed that there was also moderate to strong WTA between the farmers' preferences and the researchers' good performing varieties. The traits crucial to the farmers included resistance to pests and diseases, high yield, stiff/sturdy stem for lodging tolerance, good eating quality, tolerance to acid sulfate soils, good seed vigor, short duration, high tillering capacity, longer stalks, appropriate for existing cropping systems and less fertilizer requirement. This study employed needs-based and participatory approaches to ensure that breeding programs are well-targeted and can contribute to fast track dissemination of submergencetolerant rice varieties. The ultimate goal is increasing rice productivity and reducing poverty in flood-prone ecosystems (Ngoc Chi et al. 2011b;Mariano et al. 2012).

Kayu Agung District
Overall performance of estimated farmers' WTA at Kayu Agung was also statistically acceptable with coefficient of determination (R 2 ) of about 0.8456 and 0.8694 during the normal and flood year, respectively. In other words, about 84.56% and 86.94% of farmers' WTA variability with respect to submergence tolerant Sub-1 rice varieties was well explained by the model either during normal or flood year (Table 4).
The accuracy of estimated WTA model was also quite high with coefficient of variation (CV) 0.2355 for normal year and 0.3294 for flood year. Most of explanatory variables used in the model were significantly influenced the farmers' WTA except farmers' income during the normal year. However, during the flood year out of eight explanatory variables estimated in the WTA model only three variables that significantly influenced the farmers' WTA for the introduced submergence tolerant rice varieties such as submergence-tolerance on 14 days, high yield, and suitable rice taste.
In general, farmers showed higher response to the introduced submergence-tolerant rice varieties that expressed into WTA during the normal year compared with flood year. This indicates that farmers remain uncertain during the flood year about the opportunity of submergence tolerant rice to overcome problems at the flood prone affected rice areas in Kayu Agung.
An overall increase of indicators or explanatory variables used in the WTA model by 10% will increase farmers' WTA for submergence tolerant rice varieties in Kayu Agung by 2.4434%. This response may not as high as farmers response during the flood year, however this figure remains in an increasing trend for Kayu Agung case. Meanwhile, only few indicators that significantly affected farmers' WTA during flood year in Kayu Agung, including: (1) submergence tolerant up to 14 days, (2) high yield, and (3) suitable rice taste with estimated parameter of 0.0829, 0.1556 and 0.1118, respectively. Most of indicators did not significantly influence farmers' decision in determining their WTA for Sub-1 rice varieties during flood year, such as availability of seed, crop income in flood year, planted area in flood year, age of household's head and level of education. This response indicated that for rice cropping practices during flood year, farmers tend to use their local varieties. They remain wait for the field performance in a bigger scale.
The overall increase of 10% of each indicator during flood year will influence farmers' WTA for submergence tolerance Sub-1 rice varieties and increase by 5.3394% or more than double compared with normal year. Once again this finding showed that farmers in Kayu Agung are also more respond during the flood year compared with normal year. Need to be taken into account that farmers in Kayu Agung only grow rice once a year. Farmer faced problems and constraints during the flood year due to long flood prone submergence compared to normal year.
In addition, outputs of a study conducted in the Philippines showed that estimated factors influencing farmers willingness to accept or to adopt modern rice technology and good management practices are reasonably consistent, particularly with respect to soil deficiencies and risk aversion. Results were also consistent between models in terms of the positive impacts on the adoption of certified seed technology and integrated crop management practices of farmers' education, machinery ownership, irrigation water supply, capacity-enhancement activities and profitoriented behavior. Conversely, soil and nutrient deficiencies are impediments to their adoption. Extension-related variables have the biggest impact on technology adoption (Mariano et al. 2012).

Conclusion
Two types of submergence rice area during the wet season planting have created crop damage and significant economic loss to the farmers in Indramayu and Kayu Agung District. Meanwhile, farmer's criteria in relation to the selection, adoption and acceptance of newly introduced submergence tolerant rice varieties are location specific in nature.
Farmers are more responsive to the introduced submergence tolerant rice varieties especially during the flood year as part of their coping mechanism to the crop damage due to the flash flood or flood prone. Farmers are willing to accept or to adopt submergence tolerant Sub-1 rice varieties to reduce crop risk at flash flood rice area in Indramayu and flood prone rice affected areas in Kayu Agung as long as: 1. Seed is available when farmers need in a good quality and right amount. 2. Introduced submergence tolerant rice varieties either at flash flood or flood prone rice affected areas should more than 14 days submerge. 3. Introduced varieties produce higher yield and proffer taste such as soft texture for Indramayu and harder for Kayu Agung. 4. Higher farm households' income and planted rice areas during normal year cropping, especially wet season. 5. Age of farm household head is other factor that determines farmers' willingness to accept submergence tolerant rice varieties.

Policy Implication
1. There should be a more central government direct targeted submergence rice ecosystem and farmers group to anticipate the increasing trend of this area. 2. Expansion of irrigation canal rehabilitation program should include the most submergence affected area especially those belong to highly susceptible and nearly susceptible areas. 3. The government should focus the rehabilitation program to submergence affected rice areas due to flash flood for better drainage system. 4. Rehabilitation of up-stream area is vitally important to increase the water catchment area and reduce the flood at downstream area where rice is the main crop grown by farmers. 5. The budget is needed to construct the canal. In this regards, the local government should put the budget in the development planning for drainage construction and rehabilitation. 6. The government should consider short-term strategies that offset environmental adversities such as drought and submergence. 7. Government interventions to improve the educational status of farming households, overcome the effects of small farm size and encourage more profit-oriented behavior by farmers are necessary to enhance farmer willingness to accept or adopt introduced rice production technology particularly submergence tolerant rice varieties.