The hyper-arid western region of Rajasthan, forming a major part of the Thar Desert, is among India’s most challenging agro-ecological zones. It is characterized by extremely low and erratic rainfall (<250 mm annually), high evapotranspiration, nutrient-poor sandy soils and wide diurnal temperature fluctuations (Chauhan, 2003). These conditions limit agricultural productivity, restrict crop diversification and render farming systems largely rainfed and highly vulnerable to climatic variability (Anonymous, 2021).
       
In these fragile ecosystems, arid legumes play a crucial role in sustaining agriculture and rural livelihoods. Moth bean (Vigna aconitifolia), cluster bean/guar (Cyamopsis tetragonoloba) and green gram/moong bean (Vigna radiata) are well adapted to arid conditions due to their drought tolerance and short duration (Singh et al., 2008; Nain et al., 2015). Beyond adaptability, these crops enhance soil fertility through biological nitrogen fixation, provide dietary protein and contribute to household nutritional security (Pal et al., 2014).
       
Economically, cluster bean is significant for guar gum production, widely used in food, textile, pharmaceutical and petroleum industries, while moth bean and moong bean serve as pulses and livestock fodder (Rathore, 2001).  These crops also provide valuable fodder for livestock, which is an integral component of the mixed farming system commonly practised in arid regions. Despite their importance, productivity remains low (200-400 kg ha^-1) due to erratic rainfall, poor soils, limited irrigation, inadequate inputs and low adoption of improved agricultural technologies (Anonymous, 2021).
       
Research institutions such as CAZRI and the network of Krishi Vigyan Kendras have developed improved varieties, seed treatment methods, nutrient management and integrated pest management strategies (Anonymous, 2014). However, adoption remains limited due to technical, socio-economic and institutional constraints (Lal et al., 2015). In Jaisalmer district, agriculture is predominantly rainfed and reliant on drought-tolerant legumes, but low adoption of improved practices continues to constrain productivity (Singh and Waris, 2002).

Study area
 
The present study was conducted in Jaisalmer district of Rajasthan, India, which lies in the western part of the Thar Desert (Fig 1). The district is characterized by a hyper-arid climate with extremely low and erratic rainfall (generally less than 250 mm annually), high evapotranspiration rates, sandy soils with poor organic matter and extreme temperature variations between day and night. Agriculture in this region is predominantly rainfed and largely dependent on drought-tolerant crops such as cluster bean (Cyamopsis tetragonoloba), moth bean (Vigna aconitifolia) and green gram (Vigna radiata). These crops constitute an important component of the farming system and play a significant role in sustaining rural livelihoods in the region.


 
Sampling procedure
 
A multistage sampling technique was employed for the selection of respondents, which is commonly used in social science and agricultural extension research (Kothari, 2004). In the first stage, Jaisalmer district was purposively selected due to the prevalence of arid legume cultivation and the implementation of development programmes aimed at improving agricultural productivity. In the second stage, three villages-Basanpeer, Narshmo-ki-Dhani and deda-were selected because the scheduled caste sub Plan (SCSP) programme was implemented in these villages, where a majority of farmers belong to the scheduled caste community. In the third stage, a list of farmers involved in arid legume cultivation was prepared with the help of local extension officials. From this list, a total of 120 farmers were selected for the study using a random sampling method. The selected respondents included 60 beneficiary farmers who had received support under the SCSP programme and 60 non-beneficiary farmers who had not received such support. Both male and female farmers belonging to the scheduled caste category were included in the sample.
 
Data collection
 
Primary data were collected using a well-structured and pre-tested interview schedule. The questionnaire consisted of three major sections: (i) socio-economic characteristics of the respondents, (ii) adoption level of recommended arid legume production technologies and (iii) constraints perceived by farmers in adopting improved technologies. Personal interviews were conducted with the selected farmers to collect reliable information regarding their farming practices and technology adoption behaviour. Structured interview methods are widely recommended in agricultural extension research for collecting field-level information (Kerlinger, 1986).
 
Measurement of adoption level
 
The adoption level of recommended arid legume production technologies was measured using an adoption scale. The scale included major recommended practices such as improved varieties, seed treatment, recommended seed rate, line sowing, nutrient management, weed management, plant protection measures and irrigation practices where applicable. Each practice was assigned a score based on the level of adoption by the respondent. The adoption score was calculated by summing the scores obtained for all recommended practices. Based on the total score, the adoption level of farmers was categorized into different groups for further analysis. Similar adoption measurement approaches have been widely used in agricultural technology adoption studies (Rogers, 2003; Ray, 1991).
 
Calculation of adoption gap
 
The adoption gap was calculated to determine the difference between recommended practices and the actual practices followed by farmers. The adoption gap was computed using the following formula:
 

AG = (R-A)Rx100

 
Where,
AG= Adoption gap (percentage).
R= Recommended score for a particular practice.
A= Actual adoption score obtained by the farmer.
       
This method helped in identifying the extent to which farmers deviated from the recommended technology package and has been widely used in adoption and extension research.
 
Statistical analysis
 
The collected data were compiled, coded and analyzed using appropriate statistical tools. Descriptive statistics such as frequency, percentage, mean and standard deviation were used to analyze the data, which are commonly applied in agricultural research (Gomez and Gomez, 1984). Garrett’s ranking technique was used to prioritize constraints (Garrett and Woodworth, 1969). The adoption gap was calculated to determine the extent of deviation between recommended practices and the actual practices followed by farmers. Garrett’s ranking technique was employed to identify and prioritize the constraints faced by farmers in adopting improved technologies. In addition, the t-test was used to examine the statistical differences in adoption levels between beneficiary and non-beneficiary farmers for various recommended practices.
 
Ranking of constraints
 
To identify and prioritize the constraints faced by farmers in adopting improved arid legume technologies, Garrett’s ranking technique was employed. Respondents were asked to rank the constraints according to their importance. The ranks assigned by the farmers were converted into percent positions using the following formula (Garrett and Woodworth, 1969).
 

 

 
Where:
Rij= Rank given for the ith factor by the jth respondent.
Nj= Number of factors ranked by the jth respondent.
       
The per cent positions were then converted into garrett scores using Garrett’s conversion table. The scores for each constraint were summed and averaged to determine the mean score. The constraint with the highest mean score was considered the most significant constraint affecting technology adoption.
 

Adoption gaps in production practices
 
Analysis of adoption gaps (Table 1) showed that both beneficiary and non-beneficiary farmers had considerable gaps for most recommended arid legume production practices. Beneficiary farmers exhibited lower adoption gaps across practices, indicating the positive influence of extension interventions. Among beneficiary farmers, the largest gaps were observed for plant protection measures (88.0%), organic manure and fertilizer management (85.0%) and weed management (84.0%). Similar findings regarding technological gaps in pulse production technologies have also been reported by Babu et al., (2019). These practices require technical skills, timely inputs and financial resources, which constrain full adoption. In contrast, lower gaps were found for sowing and spacing (65.0%) and soil and field preparation (73.5%), which involve routine operations requiring fewer additional resources. The adoption of high-yielding varieties (76.0%) was moderate, suggesting awareness exists, but uptake is limited by input access. These practices are comparatively easier to implement and require fewer additional resources. Similar observations were reported by Meena et al., (2021) and Ramniwas et al., (2022), who highlighted that demonstration-based extension programmes significantly enhance farmers’ adoption of improved crop production technologies. Non-beneficiary farmers recorded higher adoption gaps across most practices. The highest gaps were in organic manure and fertiliser management (90.8%), plant protection (90.5%) and weed management (87.7%). The overall mean adoption gaps were 78.3% for beneficiary farmers and 82.0% for non-beneficiaries, underscoring the role of extension exposure in improving farmers’ adoption behaviour. Similar differences in adoption behaviour between trained and non-trained farmers have been reported by Mishra et al., (2021) in their study on adoption of improved agricultural technologies among rural farmers.


 
Distribution of adoption gap levels
 
Table 2 summarises the distribution of farmers by adoption gap category. Among beneficiary farmers, 47% were classified in the medium adoption gap group, 35% in the high and 18% in the low gap category. In contrast, the majority of non-beneficiary farmers (58%) fell within the high gap group, while only 10% were in the low gap category. Similar findings have been reported by Kanwat et al., (2011) and Meena et al., (2012), who emphasized that farmers’ knowledge level, attitude and exposure to extension services significantly influence adoption behaviour.

​
 
Constraints in technology adoption
 
Table 3 presents the major constraints reported by farmers. Capital limitations were ranked highest (MPS 89.1), reflecting poor financial capacity to procure quality seeds, fertilizers and plant protection chemicals. Marketing constraints (MPS 88.9) and poor availability of inputs (MPS 87.7) were also critical barriers. Small and fragmented landholdings (MPS 78.9) limited mechanisation and full implementation of recommended practices. Similar constraints related to socio-economic and farm characteristics influencing crop adoption have been reported by Kanwat et al., (2017) in their study on factors influencing ajwain cultivation. Labour shortages (MPS 67.9) and inadequate technical knowledge regarding agronomic practices (MPS 66.8) were additional deterrents. Infrastructural issues such as poor fencing (MPS 56.7) and inadequate storage facilities (MPS 56.8), along with environmental challenges like drought (MPS 44.9), further constrained adoption. These findings suggest that economic, institutional and infrastructural factors jointly influence technology adoption in arid regions. Similar economic constraints affecting adoption of improved agricultural technologies have been reported by Chahal and Kataria (2010); Ranawat and Ram (2015) and Vadivelu and Kiran (2013). Input availability and marketing constraints were also major barriers in adoption, which is consistent with the findings of Srikanth et al., (2020) and Choudhary et al., (2025).


 
Multi-crop adoption gap analysis
 
Table 4 provides crop-wise adoption gaps for different legume species. In chickpea, the highest gaps were observed in harvesting (85.0%) and organic manure and fertilizer management (81.0%). For moongbean, weed management (84.0%) and plant protection (82.0%) showed the largest gaps. Similar patterns were seen in mothbean where weed management (87.0%) and harvesting (83.0%) were major gaps and in clusterbean where plant protection measures (88.0%) recorded the highest gap. The consistent pattern of high gaps in plant protection, nutrient and weed management across crops highlights areas where focused extension and input support are most needed.


 
Comparison between beneficiary and non-beneficiary farmers
 
Statistical comparisons (Table 5) revealed significant differences in several practices. Beneficiary farmers had significantly higher adoption in weed management (t=3.90) and harvesting (t=3.45). Differences were also significant for adoption of high-yielding varieties (t=3.21), organic manure and fertilizer management (t=2.85) and plant protection (t=2.31). However, there were no significant differences in soil and field preparation, sowing and spacing, seed treatment and storage, indicating that these practices are routinely followed independent of extension exposure.


 
Overall adoption levels and technological gaps
 
Table 6 shows the average adoption levels of different practices. Sowing and spacing recorded the highest mean adoption (46.9%), followed by soil and field preparation (44.9%) and weed management (43.3%). Practices such as high-yielding variety adoption (25.0%) and storage (24.5%) showed the lowest adoption, resulting in the highest technological gaps (75.0% and 75.5%, respectively). Substantial gaps were also observed in plant protection measures (69.5%), seed treatment (66.8%) and harvesting (64.6%). Variation in adoption behaviour suggests heterogeneity in resource access and knowledge among farmers.

The study shows substantial adoption gaps in improved arid legume production practices. Beneficiary farmers had a lower overall adoption gap (78.3%) compared to non-beneficiaries (82.0%), indicating that extension services play an important role in reducing technological gaps. The largest gaps were noted in plant protection, nutrient management and weed control, while routine operations such as sowing and soil preparation recorded smaller gaps. Distribution of farmers by adoption gap category highlighted that extension exposure shifts farmers toward lower gap levels. Key constraints such as limited capital, marketing difficulties and restricted access to quality inputs were identified as primary barriers. Addressing these through strengthened extension programmes, improved input supply and market support will be critical to enhance adoption. Reducing the technological gap in arid legume production is essential for increasing productivity, boosting farm income and fostering sustainable agricultural development in arid and semi arid regions.

The authors gratefully acknowledge the Director, ICAR-Central Arid Zone Research Institute, Jodhpur, for financial support and research facilities. Thanks are also due to ICAR Regional Research Stations, Jaisalmer and Bhuj, for technical assistance in data compilation and documentation.
 
Disclaimers
 
The views and conclusions expressed in this article are solely those of the authors and do not necessarily represent the views of their affiliated institutions. The authors are responsible for the accuracy and completeness of the information provided, but do not accept any liability for any direct or indirect losses resulting from the use of this content.
 
Informed consent
 
All the experimental procedures were approved by the Institute’s research committee.

The authors declare that there is no conflict of interest regarding the publication of this article. No funding or sponsorship influenced the design of the study, data collection, analysis, decision to publish, or preparation of the manuscript.