An Appraisal of the Constraints, Opportunities, and Farmers’ Needs and Preferences of Oil Palm for Sustainable Production and Improvement in Tanzania

Abstract

Oil palm is one of the primary vegetable oil sources worldwide, including in Tanzania. Tanzania’s mean palm oil yield is 1.6 tons per hectare, far below the 6 to 8 tons per hectare reported elsewhere. This low oil yield is attributable to underdeveloped, unsustainable oil palm production systems and improvements, several biotic and abiotic stresses, and socio-economic and policy challenges that have yet to be systematically documented to guide large-scale production, breeding, and research support. The objectives of this study were to appraise oil palm production and improvement in Tanzania, focusing on constraints, opportunities, and farmers’ major preferences. A participatory rural appraisal study was conducted in Kigoma Region, in three selected districts. Data were collected from 392 oil palm farmers using semi-structured questionnaires and 54 focus group discussants. Data were subjected to statistical analyses to discern the variables and their significant associations using the Statistical Package for Social Science (SPSS Inc., 2020). About 98.5% of the participant farmers engage in oil palm production. Most respondent farmers predominantly cultivate the Dura oil palm type (97.4%), followed by Tenera (50%). The farmers’ major reported oil palm production constraints were an inadequate supply of improved planting materials (reported by 82.7% of respondents), poor access to credit (72.4%), a high cost of production inputs (59.4%), poor market access (56.4%), insect pests and diseases (53.6), and poor production technologies (45.4%). A chi-square analysis of farmers’ production constraints revealed that the unavailability of labor (X² = 41.181; p = 0.000); limited extension services (X² = 29.074; p = 0.000); and diseases and pests (X² = 19.582; p = 0.000) differed significantly across the study area. Additionally, the lack of fertilizers (X² = 14.218; p = 0.001); inappropriate technology and knowledge gaps (X² = 10.529; p = 0.005); and poor market access (X² = 6.621; p = 0.036) differed significantly across districts. A high oil yield (reported by 58.7% of the respondents), a high number of bunches per plant (40.5%), early maturity (37.2%), and tolerance to droughts (23%) and diseases and insect pests (18.9%) were the most preferred traits by farmers in oil palm varieties. Therefore, integrative and sustainable breeding oil palm for enhanced yields and farmers’ preferred traits will increase the adoption of newly improved varieties for local palm oil production, import substitution, and economic development in Tanzania.

1. Introduction

Oil palm (Elaeis guineensis Jacq., 2n = 2x = 32) is one of the main vegetable oil sources worldwide, including in Tanzania. The genus Elaeis belongs to the palm family Palmae and order Spadiciflorae, an important monocot member. The word Elaeis originates from the Greek word Elaion, meaning oil, and guineensis points to the oil palm’s origin on the Guinea coast in Africa [1]. Oil palm has the highest oil yields per hectare compared to other oil-producing crops, such as sunflower, cottonseed, peanut, soybeans, sesame, and rapeseed [2]. The crop requires 1500 to 3000 mm of rainfall and 25 to 32 °C temperatures for optimum growth and production. Indonesia and Malaysia dominate global production, while regions in Sub-Saharan Africa and parts of Southern Asia have also developed oil palm industries [3,4,5]. In Tanzania, oil palm holds considerable potential to boost the local economy, reduce the reliance on imported edible oils, and provide livelihoods for farmers. However, despite this potential, oil palm production and yields in Tanzania have remained stagnant, contributing only 15% of the total production [6]. The remaining 85% of edible oil sources are from other oil crops, including sunflower (Helianthus annuus L.), groundnut (Arachis hypogaea L.), cashew nut (Anacardium occidentale L.), and sesame (Sesamum indicum L.). Oil palm productivity in the country is 1.6 tons per hectare, lower than the 5 to 8 tons per hectare reported elsewhere [7]. The current total production of edible oil is 290,000 tons per annum, about 40% of the total annual demand, with the rest being met through imports. The low productivity in the country is attributed to the use of low-yielding oil palm varieties, poor agronomic practices, lack of modern production technologies, lack of extension services, lack of modern processing technologies, and [8] low-quality planting material. These challenges are compounded by the ageing plantations (90% of which are old and not replanted), resulting in low yields, far below the global average.

Although several production challenges have been identified, there is a lack of detailed and comprehensive studies that address the specific barriers farmers face in Tanzania, especially on how these constraints vary across regions, farm sizes, and different degrees of market access. Furthermore, there is insufficient understanding of the preferences and needs of farmers regarding the development of improved hybrid oil palm varieties that meet local production and market requirements. The Government of Tanzania has prioritized oil palm crops to boost the local production of palm oil for sustainable edible oil supply and marketing. The renewed initiative led to the establishment of the Kihinga Research Center under the Tanzania Agricultural Research Institute (TARI) in 2018, which mandated conducting and coordinating oil palm research and development in the country. The institute has started the production of Tenera seeds by crossing selected Dura × Pisifera parents, aiming at developing new genetic resources and to deploying better-performing selections and seedlings to farmers. Yet, some growers are dependent on farmers’ saved and recycled planting materials from Dura types due to several challenges that are yet to be systematically addressed to guide large-scale production, breeding, and policy support.

The current research on oil palm in Tanzania has primarily focused on deciphering general production constraints. However, it has not adequately addressed how these constraints interact or cumulatively affect sustainable production and improvement [8,9]. There is a significant gap in understanding how farmers’ knowledge, preferences, and practices align with or diverge from the new breeding initiatives and agronomical and production technologies. Additionally, no recent studies have investigated the impact of the entire oil palm value chain, from the availability of planting materials to market access. Participatory rural appraisal (PRA) is the multidisciplinary research approach that helps identify production constraints that focus on farmers’ needs and value chains [10]. Previous PRA studies were conducted to initiate oil palm research programs and develop policies to optimize oil palm production and improve farmers’ livelihoods in different agro-ecologies [11,12,13,14]. For instance, a PRA study was conducted in Benin by Akpo, Vissoh [11], who reported a poor genetic quality of the planting materials, the poor allocation and geographical distribution of nursery sites, the high cost of hybrid planting materials, poor seedling care in nurseries, leading to poor yields and quality gains. De Vos and Delabre [13] reported unequal access to land and resources and a lack of participation by women during negotiations on land acquisition and the production of oil palm in West Kalimantan, Indonesia. Additionally, in Ethiopia, Teklu, Shimelis [15] reported a lack of access to improved seeds, diseases, the high cost of seeds, and the low market price as key challenges for sesame production. This study seeks to address these gaps by providing a detailed assessment of the current oil palm production system in Tanzania, focusing on the constraints, opportunities, and the specific needs and preferences of farmers to guide sustainable production and breeding. Specifically, this study analyzes the main production challenges and opportunities along the oil palm value chain and outlines drivers that influence the continued use of locally un-improved oil palm varieties. Hence, the objectives of this study were to appraise oil palm for sustainable production and improvement in Tanzania, focusing on constraints, opportunities, and farmers’ major preferences. The findings may inform integrative and sustainable breeding oil palm for enhanced yields and farmers’ preferred traits and may increase the adoption of newly improved varieties for local palm oil production, import substitution, and economic development in Tanzania.

2. Materials and Methods

2.1. Description of the Study Sites

The study was conducted in the Kigoma region, which is the leading oil palm-producing region in Tanzania. The region is located in western Tanzania on the shores of Lake Tanganyika, between 3.6° and 6.5° south and longitude 29.5° and 30.5° east (Figure 1). The region contributes 73% to the total oil palm production in the country [6]. The region has eight district councils that are involved in oil palm production.

Table 1. Agro-climatic characteristics of the study area in the Kigoma region of Tanzania.

Districts	Altitude (Masl)	Latitude	Longitude	Annual Rainfall (mm)	Temp Average (°C)	Relative Humidity (%)	Wind Speed (m/s)
Kigoma Urban	786	−4.83571	29.67765	1112.7	23.8	75.67	2.77
Kigoma Rural	920	−4.78422	29.76672	1087	22,5	85.81	2.40
Uvinza	1087	−5.21104	29.84361	986	21.3	89.05	2.51

Note: Masl = meter above sea level; mm = millimeter; m/s = meter per second; °C = Celsius.

presents the details of the three districts, including Kigoma Urban, Kigoma Rural, and Uvinza, selected as major producers of oil palm in the region.

The districts are characterized by a tropical climate with unimodal rainfall from late October to May. During the study period, the mean annual rainfall in the area ranged from 986 mm to 1112.7 mm. Daily mean temperatures ranged between 21.3 °C and 23.8 °C and varied with altitude. The soils are sand loamy, deep, and well-drained, especially in Kigoma Urban. In high-altitude areas, soils are black and brown alluvial, while at low altitudes, soils are dark red loams with good drainage. Relative humidity ranged from 75.67 to 89.05%, whereas wind speed ranged from 2.40 to 2.77 m per second.

2.2. Sampling Method

The Kigoma region was purposively selected for this study, given its high levels of oil palm production. The Uvinza, Kigoma Rural, and Kigoma Urban district councils are the major palm oil producers in the region and were selected for this study [6]. These districts contribute to 86.8% of the total production in the region, with the highest contribution from Uvinza (34.8%), followed by Kigoma Rural (32.0%) and Kigoma Urban (19.9%) for the 2023/2024 season (Figure 2). The three districts are represented by wards and villages, as presented in

Table 2. Description of the study sites.

Districts	Wards	Latitude	Longitude	Altitude	Number of Sampled Farmers	Villages
Uvinza	Ilagala	−5.162811	29.826670	764.6	20	Ilagala, Sambara
	Kandaga	−4.961684	29.844719	1005.6	58	Kandaga, Kalenge, Mlela
	Mwakizega	−5.094654	29.814871	805.2	10	Mwakizega, Kabeba
Kigoma Rural	Bitale	−4.749048	29.694431	988.1	44	Kizenga, Bitale
	Mahembe	−4.809049	29.735140	958.0	28	Mahembe, Chakabwimba
	Mungonya	−4.851562	29.695223	834.4	28	Msimba, Kamala
	Simbo	−4.882163	29.740887	808.7	25	Simbo, Matiazo, Kasuku
	Mkongolo	−4.693834	29.714690	1062.6	22	Mkongolo, Nyamhozya
	Mwandiga	−4.839023	29.661537	820.4	44	Mwandiga, Kibingo, Bigabiro, Kiganza
	Nkungwe	−4.822669	29.790406	880.8	20	Nkungwe, Bigere
Kigoma Urban	Businde	−4.884062	29.699048	779.2	26	Mungonya, Msufini
	Kagera	−4.884011	29.698896	775.6	67	Mgumile, Kagera, Kanswa

. The 29 villages were selected using a probability proportional to size (PPS) method from each selected district [16]. A systematic random sampling procedure (SRS) was used to select smallholder oil palm farming households. The sample size was determined using Krejcie and Morgan’s [17] method, as shown in Section 2.3.

2.3. Sample Size Determination

According to the National Statistical Bureau [6], 29,255 households were involved in oil palm production in the 2021/2022 cropping season. This study adopted the method reported by Krejcie and Morgan [17] to determine an adequate sample size. This method considers a sample size to provide accurate statistical estimates with a 5% sampling error and allows for comparisons and stratifications for significant associates. The following equation was used to derive the appropriate sample size for the survey.

$$n={\displaystyle \frac{x^{2}NP(1-P)}{d^2(N-1)+x^{2}P(1-P)}}$$

(1)

• where
• n = the required sample size; $x^2$ = the table value of chi-square for 1 degree of freedom at the desired confidence level (3.841), i.e., (1.96 × 1.96 = 3.8416), with a 95% confidence level (standard value is 1.96); N = the population size; P = the population proportion/variance in the population (that is set at 0.50 to provide maximum sample size); and $d$ = the degree of accuracy/margin of error at 5% (standard value of 0.05).

$$n={\displaystyle \frac{3.8416\ast \mathrm{29,255}\ast 0.5\ast 0.5}{{0.05}^2(\mathrm{29,255}-1)+3.8416\ast 0.5\ast 0.5}}=379.1936\approx 380$$

(2)

From the determined sample size (380), 5% was added to cater for non-responses, as proposed by Krejcie and Morgan [18], producing a sample size of 399. During data collection, the sample size covered was 392 smallholder oil palm farmers, resulting in a 1.75% missing data rate, which was random. Given the low percentage of non-responses, no significant bias was anticipated in the results. A listwise deletion method was employed, where cases with missing values were excluded from the analysis.

2.4. Data Collection

Quantitative data were collected from sampled smallholder oil palm farming households using a semi-structured questionnaire. The questionnaires were coded using a Microsoft Excel spreadsheet and were uploaded to the Kobo Collect Server (Harvard Humanitarian Initiative, Cambridge, MA, USA) for data collection. The questionnaire was programmed into Kobo Collect mobile data collection, where the validation rules, including skip pattern rules, mandatory rules, GPS locations, drop-down menus, ranging rules, and the logic of numbers between one variable and another, were created in the system. Using mobile data collection tools minimizes data entry errors and monitors data collection.

With the help of agricultural extension officers, data were collected from the selected farming households at their households or farms or workplaces. Questionnaires were pre-tested to evaluate the clarity, reliability, and validity of a relatively small sample and to identify potential issues before the main survey. The main survey activities included introducing the research objective, verifying eligibility, obtaining informed consent, and conducting face-to-face interviews. The questionnaire’s contents aimed at capturing data on various aspects of the socio-economic status of the households, knowledge and trainings, production constraints, crop management practices, access to resources, and farmers’ perceptions and preferences.

Focus Group Discussions (FGDs) and Key Informant Interviews (KIIs) were used to collect qualitative data. One FGD was held in each of the selected districts, with each group comprising at least 18 participants. The FGD predominantly comprised open-ended and deep probing questions. In addition, tablets and voice recorders were used to collect FGD participants’ voices, photographs, and GPS locations, which were stored in the cloud server for reference and backup. Key informants’ interviews involved in-depth interviews conducted with selected knowledgeable individuals in the oil palm sector and guided by a checklist of questions for data collection. A total of 54 farmers (18 from each district) participated both in FGDs and KIIs. Selected individuals were experienced farmers and community leaders known for their rich indigenous and technical knowledge of oil palm production. Data quality checks included data checking and cleaning to produce reliable data files. The cleaning involved detecting and correcting inconsistencies and outliers in the data, since extreme values are major contributors to sampling variability in survey estimates.

2.5. Data Analysis

Quantitative data collected through questionnaires were downloaded from the server, saved in MS Excel, and then exported to the Statistical Package for Social Science (SPSS Inc., Chicago, IL, USA, 2020) for alignment with the code book and value labels. Statistical analyses were performed to summarize the category of the variables and their significant associations. The chi-square goodness-of-fit statistic was computed using significance tests to discern associations among the variables. The descriptive statistical analysis included proportions, frequencies, percentages, tabulations, and cross-tabulations of critical variables. This allowed for empirical analyses and descriptions of associations between the collected parameters across the study districts.

2.6. Limitations of This Study

This study primarily focuses on diagnosing the constraints, opportunities, and farmers’ needs and preferences of oil palm for sustainable production and improvement in Tanzania. It emphasizes access to improved and quality seedlings; seedling supply systems across broader value chains, such as the establishment of seedling nurseries and plantations; marketing; and processing, which could also influence the future of oil palm production in Tanzania. The identified limitations suggest areas for further research and development. Also, there is a need for a quantitative analysis of the diagnosed constraints to guide policies and for understanding their impact in the oil palm industry.

3. Results

3.1. Description of Households

A total of 392 smallholder farmers were interviewed during the study. All respondents indicated that they were involved in oil palm production. The results show that 82.4% of the interviewed oil palm growers were males, while 17.6% were females (

Table 3. Demographic profile of the interviewed oil palm growing farmers in Tanzania.

Variables	Category	Districts			Overall (n = 392)	Chi-Square Statistics
		Uvinza (n = 88)	Kigoma Rural (n = 211)	Kigoma Urban (n = 93)		df	Chi-Square Value	p-Value
Gender	Male	74	181	68	323	2	7.362	0.025
	Female	14	30	25	69
Education level	Primary	76	177	61	314	6	25.087	0.000
	Secondary	1	16	11	28
	College	1	3	0	4
	Illiterate	10	15	21	46
Age	<35 years	2	8	7	17	2	3.34	0.189
	>35 years	86	203	86	375

Note: n = total number of samples, df = degree of freedom.

). The proportion of sampled males and females exhibited a similar trend across the study. The Pearson chi-square results reveal that gender and location of residence (district) significantly affected whether an oil palm farmer would be male or female (X² = 7.36; p = 0.025). Education level significantly differed across the study area (X² = 25.087; p = 0.000). Most respondents had completed primary school (80%), and a few had reached secondary school (7%). Conversely, 1% of the respondents had reached tertiary education, while 11.7% did not achieve formal education. Also, the age of respondents did not differ significantly among the studied districts (X² = 3.34; p = 0.189).

Overall, 95.6% of the interviewed farmers were adults above 35 years old and are able to make decisions on crop cultivation. Among the respondents, those with an average age of 56 years had experience of about 24 years in oil palm farming activities. Other variable features of smallholder respondent farmers existed in the study area, including farming experience, farm size, and the number of oil palm trees owned by the farmer’s household. On average, a farmer had small land holdings of 1.45 hectares with 222 oil palm trees. The results show a mean of 153 oil palm tree populations per hectare, above the recommended 123 oil palm trees in a rectangular or 142 in a triangular planting pattern at an inter- and inter-row spacing of 9 m × 9 m.

3.2. Socio-Economic Activities

The main economic activities identified in the study area were crop and livestock production, trade, and hired workers (

Table 4. Socio-economics of and income generated in the study area.

Primary Occupation	Districts			Overall (n = 392)	Chi-Square Statistics
	Uvinza (n = 88)	Kigoma Rural (n = 211)	Kigoma Urban (n = 93)		df	Chi-Square Value	p-Value
Crop farming	86	201	90	377	6	5.36	0.499
Livestock production	0	0	1	1
Trade	2	9	2	13
Hired worker	0	1	0	1

). There were non-significant (X² = 5.36; p = 0.499) differences in the respondent farmers engaged in different economic activities in the study area. About 96% of the respondents were involved in crop production as the main socio-economic activity, indicating a high contribution to community livelihood compared with trade (3%), livestock production (<1%), and hired workers (<1%). When comparing income generated from different enterprises, oil palm production had the highest annual income for smallholder farmers across the districts. Each household earned an average income of about TZS 2.8 million yearly (Figure 3). This is attributed to 98.5% of the farmers being involved in oil palm production activities. About 45.9% of the respondent farmers earned income from selling other produce, especially grain maize and beans and cassava roots. However, when compared to other crops, oil palm offers high potential returns due to its higher yields and market value. Additionally, this income is above the national poverty threshold, which is approximately TZS 5698.98 (USD 2.15). The average household income from the sale of the produce was TZS 418,956. Livestock, vegetable and fruit production, and remittances provided an income of 10%. Only 4% of the respondent farmers earned income from salaries and wages. Their average annual income from salaries and wages was TZS 2.3 million. Therefore, oil palm offers a relatively higher income compared to other activities, providing a potentially more sustainable livelihood for farmers.

3.3. Oil Palm Production Constraints

Farmers’ major oil palm production constraints were an inadequate supply of improved planting materials (reported by 82.7% of the respondents), poor access to credit (72.4%), a high cost of production inputs (59.4%), poor market access (56.4%), insect pests and diseases (53.6%), and backward production technologies and knowledge gaps (45.4%) (

Table 5. Major oil palm production constraints in the study area.

Production Constraints	Category	District			Overall (n = 392)	Chi-Square Statistics
		Uvinza (n = 88)	Kigoma Rural (n = 211)	Kigoma Urban (n = 93)		df	Chi-Square	p-Value
Lack of improved plating material	Yes	79	174	71	324	2	5.698	0.058
	No	9	37	22	68
Lack of suitable production technologies	Yes	53	84	41	178	2	10.529	0.005
	No	35	127	52	214
Poor soil health	Yes	19	49	15	83	2	1.958	0.376
	No	69	162	78	309
Lack of capital	Yes	60	159	65	284	2	2.000	0.368
	No	28	52	28	108
Lack of fertilizers	Yes	48	112	29	189	2	14.218	0.001
	No	40	99	64	203
Unavailability of labor	Yes	55	80	15	150	2	41.181	0.000
	No	33	131	78	242
Erratic rainfall	Yes	9	32	9	50	2	2.399	0.301
	No	79	179	84	342
Diseases and insect pests	Yes	62	113	35	210	2	19.582	0.000
	No	26	98	58	182
Poor market access	Yes	60	110	51	221	2	6.621	0.036
	No	28	101	42	171
Limited extension service	Yes	50	58	22	130	2	29.074	0.000
	No	38	153	71	262
High cost of production inputs	Yes	59	126	48	233	2	4.481	0.106
	No	29	85	45	159
Limited land size	Yes	23	34	23	80	2	5.243	0.073
	No	65	177	70	312
Inadequate irrigation water	Yes	18	25	21	64	2	3.826	0.148
	No	70	177	82	329

Note: n = total number of samples; df = degree of freedom.

). On the other hand, the least ranked constraints were limited labor availability (38.3%), limited extension services (33.2%), soil infertility problems (21.2%), limited land sizes (20.4%), inadequate water for irrigation (13.8%), and erratic weather patterns (12.8%). The chi-square analysis revealed that the unavailability of labor (X² = 41.181; p = 0.000); limited extension services (X² = 29.074; p = 0.000); and diseases and pests (X² = 19.582; p = 0.000) differed significantly across the study area. Additionally, a lack of fertilizers (X² = 14.218; p = 0.001); inappropriate technology and knowledge gaps (X² = 10.529; p = 0.005); and poor market access (X² = 6.621; p = 0.036) differed significantly across districts.

3.4. Access to Extension Service and Improved Oil Palm Seedlings

Extension services are of paramount importance to oil palm farmers. Extension officers link researchers and farmers, guiding good crop management practices. In the study area, 31.8%, 37.9%, and 31.2% out of 88, 211, and 93 of the interviewed oil palm farmers in Uvinza, Kigoma Rural, and Kigoma Urban benefited from extension services in the 2021/22 planting season, respectively. About 35% out of the 392 respondents benefited from extension services. The chi-square analysis revealed that there were non-significant differences between districts on access to extension services (X² = 1.776; p = 0.441) and access to improved varieties of oil palm seedlings (X² = 5.580; p = 0.061) (

Table 6. Summary statistics on access to extension services and production technologies for oil palm production in the study area.

Access to Services and Technologies	Category	District			Overall (n = 392)	Chi-Square Statistics
		Uvinza (n = 88)	Kigoma Rural (n = 211)	Kigoma Urban (n = 93)		df	Chi-Square	p-Value
Extension services	Yes	28	80	29	137	2	1.776	0.411
	No	60	131	64	255
Fertilizers	Yes	6	24	5	35	2	3.477	0.176
	No	82	187	88	357
Planting materials (seedlings)	Yes	18	72	27	117	2	5.580	0.061
	No	70	139	66	275

). Field visits by extension officers, training offered by TARI staff, various popular articles published in newspapers, radio, television, and annual agricultural exhibitions (locally referred to as Nane Nane) were among the primary sources of information for the crop growers. These sources enabled some farmers to make decisions, including using improved oil palm varieties, fertilizers, and good agronomic practices, replacing old oil palm trees, accessing new agricultural lands, and forming an Agricultural Marketing Cooperative Society (AMCOS). However, most of the interviewed extension officers admitted having inadequate knowledge of oil palm production and management practices due to a lack of relevant information on oil palm production for farmers, extension officers, and other stakeholders.

3.5. Use of Fertilizers for Oil Palm Production

In the study area, only 8.9% of respondents reported using organic fertilizer (Figure 4). The number of farmers who do not use chemical fertilizers was the highest in Kigoma Rural (11.4%) compared to Kigoma Urban (5.4%) and Uvinza (6.8%) districts. The chi-square statistical analysis indicated a non-significant (X² = 3.477; p = 0.176) difference in the use of fertilizers across districts (Table 6). Most respondent farmers (80%) used farmyard manure, while Nitrogen:Phosphorus:Potassium (NPK), Calcium Ammonium Nitrate (CAN), Urea, Sulphate Ammonium (SA), and Minjingu Rock Phosphate (MRP) were used by only 20% of the respondent farmers. The fertilizers used by the farmers were intended for intercrops (e.g., maize, common beans, and groundnuts) but not those of oil palm.

3.6. Weed Management in Oil Palm Production

Most of the oil palm farmers (96%) engage in weed management of their oil palm fields (

Table 7. Weeding practices among oil palm farmers in the study area.

Variable	Category	District			Overall (n = 392)	Chi-Square Statistics
		Uvinza (n = 88)	Kigoma Rural (n = 211)	Kigoma Urban (n = 93)		df	Chi-Square	p-Value
Weed management	Yes	96.59	96.21	95.7	96.17	2	0.990	0.952
	No	3.4	3.8	4.3	3.8
Frequency of weeding/year		1.33	1.45	1.52	1.44	8	9.531	0.324
Weeding method
Hand hoeing	Yes	94.1	94.1	89.9	93.1	2	1.729	0.421
	No	5.9	5.9	10.1	6.9
Slashing	Yes	44.7	49.8	67.4	52.8	2	10.619	0.005
	No	55.3	50.2	32.6	47.2
Chemicals	Yes	0.3	0.0	0.0	0.3	2	3.444	0.179
	No	98.8	100	100	99.7
Burning	Yes	1.2	0.1	0.0	0.8	2	1.644	0.441
	No	98.8	99	100	99.2
Cover crops	Yes	0.0	0.5	0.0	0.3	2	1.240	0.538
	No	100	99.5	100	99.7
Biological control	Yes	0.0	0.5	0.0	0.3	2	1.358	0.507
	No	100	99.5	98.9	99.7

Note: n = total number of samples; df = degree of freedom.

). Weeding is integrated into oil palm fields intercropped with annual crops. The main weeding methods include hand hoeing, slashing, burning, biological control, herbicides, and cover crops, as reported by 93%, 52.8%, 0.8%, 0.5%, 0.3%, and 0.3% of the respondents, respectively. Use of chemical herbicides was reported only in the Uvinza district, whereas burning was reported in the Uvinza and Kigoma Rural districts. The chi-square test analysis showed that the number of farmers who weed (X² = 0.99; p = 0.952) and the frequency of weeding per year (X² = 9.531; p = 0.324) were not significantly different across the study area. In addition, the different methods of weeding, such as i.e., hand hoeing (X² = 1.729; p = 0.421); chemicals (X² = 3.444; p = 0.179); burning (X² = 1.644; p = 0.441); cover crops (X² = 1.240; p = 0.538); and biological control (X² = 1.358; p = 0.507) did not differ significantly across districts, except for slashing (X² = 10.619; p = 0.005).

3.7. Intercropping of Oil Palm with Major Annual Crops

Some 50% of the respondent farmers reported intercropping of oil palm trees with other major annual crops, such as maize, cassava, and common beans (

Table 8. Types of oil palm cultivated by smallholder oil palm farmers in the study area.

Types	District			Overall (n = 392)	Chi-Square Test
	Uvinza (n = 88)	Kigoma Rural (n = 211)	Kigoma Urban (n = 93)		df	Chi-Square	p-Value
Dura	41	92	45	178
Tenera	34	84	36	154	4	7.325	0.835
Pisifera	15	27	18	60

Note: n = total number of samples; df = degree of freedom.

). However, intercropping has advantages and disadvantages depending on the age of the oil palm and the type of crop intercropped. Some intercropping was widely practiced in the Kigoma Rural (reported by 61.1% of the respondents) compared to Kigoma Urban (51.6%) and Uvinza (45.5%) districts (Table 8). The major crops intercropped with oil palm were maize (reported by 35.7% of the respondents), cassava (26.8%), and common beans (24.7%) (Table 8). Annual crops are widely intercropped, mainly at the early oil palm growth stages (1 to 7 years after planting), which is the tree establishment phase. Other crops intercropped with oil palm were pigeon pea (reported by 10.5% of the respondents), groundnut (10.5%), sweet potatoes (5.4%), cocoyam (8.9%), banana (5.6%), pineapple (1.3%), pawpaw (1.3%), cowpea (3.1%), passion fruit (1.0%) and avocado (0.5%). The chi-square test analysis revealed that among the intercrops, the use of cocoyam (X² = 27.773; p = 0.000); banana (X² = 10.040; p = 0.000); cassava (X² = 13.665; p = 0.001); maize (X² = 10.690; p = 0.005); and passion fruit (X² = 6.350; p = 0.042) was significantly different across districts. There were non-significant differences for other remaining intercrops, such as sweet potatoes (X² = 0.476; p = 0.788); pineapple (X² = 1.432; p = 0.489); cowpea (X² = 0.640; p = 0.969); pawpaw (X² = 1.432; p = 0.489); and avocado (X² = 1.051; p = 0.591). However, intercropping was mainly practiced in Kigoma Rural and Kigoma Urban compared to the Uvinza district.

3.8. Oil Palm Types Cultivated in the Study Area

The Dura type was cultivated by most interviewed farmers (97.4%) in the study districts. However, more than 50% of the oil palm farmers had grown the Tenera type. Conversely, the Pisifera oil palm was cultivated by a limited number (2%) of respondents across the studied areas. The chi-square test analysis revealed non-significant (X² = 7.325; p = 0.835) differences among the farmers who cultivated Dura, Tenera, and Pisifera types across districts (

Table 9. Percentage of farmers’ traits of preference in oil palm production in the study area.

Traits	Districts			Overall (n = 392)	Chi-Square Test
	Uvinza (n = 88)	Kigoma Rural (n = 211)	Kigoma Urban (n = 93)		df	Chi-Square	p = Value
High oil content	58.3	63	55	58.7
Short stem height	33.8	42	36	37.2
Drought tolerance	19.8	27.1	22	23	8	6.721	0.861
High number of bunches per plant	35	47.4	39	40.5
Diseases and insect pest resistance	16	23.3	17.2	18.9

Note: X² = chi-square test; df = degree of freedom.

). No institute or company reported to breed or import the Pisifera type for production or research purposes. The Pisifera type appears by chance in Dura and Tenera fields. Farmers do not cultivate Pisifera types, because they are female-sterile and are only used by breeders as male parents during Tenera hybrid seed production.

3.9. Farmers Preferred Traits of Oil Palm Types

Through focused group discussions, farmers across the study area voted for their ideal oil palm types for commercial production (Table 9). The main reported traits of preferences identified were a high oil content and oil yield (reported by 58.7% of the respondents), a high number of bunches per plant (40.5%), early maturity (37.2%), drought tolerance (23%), and disease and pest resistance (18.9%). The chi-square test analysis revealed that the farmers’ preferred traits did not differ significantly (X² = 7.621; p = 0.861) across districts. A high oil content and many bunches per plant were commercially important traits across districts. Short stem height eases bunch harvesting. The overall ranks through FGD indicated disease and pest resistance as the least rated traits. Fungal diseases, such as ganoderma (Ganoderma boninense), and insect pests known as rhinoceros beetles (Oryctes rhinoceros) were the most common biotic constraints reported in Kigoma Rural and Kigoma Urban, respectively.

4. Discussion

4.1. Description of Households

The results of this study show that most of the respondents (95.6%) cultivating oil palm fields were adults (above 35 years old), while 4.34% were youth (>35 years old). This indicates that experienced people were mostly undertaking oil palm farming activities. These results are further supported by Mwatawala, Maguta [8], who reported that a majority (74.1%) of respondents in Kigoma district had experience of more than 10 years in the cultivation of oil palm, regardless of the type of technology adopted. However, they further reported that the age of farmers and their low level of education and experience have a negative correlation with the adoption of improved oil palm varieties. This explains that as a farmer grows older, there is decreased interest in long-term investments in farm activities, while on the other hand, younger farmers are risk-takers willing to try new technologies. These findings are further supported by Yaseen, Thapa [19], who found that age negatively influenced the adoption of oil palm cultivation technologies in Northeast Thailand. The low level of education (80% attained primary education) and small farm size (average of 1.45 hectare) reported in this study negatively influenced the adoption of improved oil palm technologies. These findings agree with those of a study by Negera, Alemu [20], which revealed that education levels influence the adoption of new agricultural technologies. This suggests that more educated farmers are likely to have high access to information and stay informed about the potential innovation, sustainability, and sense of control over new practices. Similar to farm size, a farmer with a large area is better able to administer an innovative process, because they have more funds, workers, and planning of the work to be conducted [20,21]. Conversely, some of the literature cited in this study agrees that small and medium-sized farms have always adapted and innovated to remain competitive in the market [22,23,24].

4.2. Crop Management Practices

In Tanzania, most oil palm land is owned by smallholder farmers and a few government facilities in Kwitanga and Kimbiji prisons in Kigoma and Dar es Salaam, respectively. Poorly managed plants with low productivity characterize the fields. High yields were reported by large-scale farmers whose fields are equipped with adequate infrastructures for weeding, fertilization, harvesting, disease and pest management, postharvest loss management, and basic amenities to attract plantation workers [25]. This is contrary to the interviewed farmers who reported a low use of fertilizers, low frequencies of weeding, and poor knowledge of oil palm crop production, intercropping techniques, and postharvest loss management. For example, our study revealed a plant population of 153 trees per hectare, which is higher than the recommended pattern for oil palm cultivation. This may affect yields, soil health, and disease susceptibility [26,27,28]. Past studies documented that a higher tree density may lead to increased competition for nutrients, may impact soil health, and may create favorable conditions for the growth of pathogens and the spread of diseases [29,30].

4.3. Fertilizer Application

The use of fertilizers was erratic and low in Uvinza, Kigoma Urban, and Kigoma Rural. Only 9% of the farmers applied fertilizers. The Ministry of Agriculture has subsidized effective fertilizers since 15 August 2022. However, the high cost of fertilizers, their unavailability, and the lack of technical knowledge on its importance to soil nutrient improvement were identified as the main reasons for their limited use in oil palm production. Enhancing oil palm yields hinges on good and improved planting material and best agronomic management practices, especially for efficient and optimal fertilizer use [31,32,33]. The quantity of fertilizer applied is influenced by farming practices, soil fertility, and farmers’ knowledge of application rates [32]. This study revealed that fertilizer management on oil palm farms is often based on traditional methods rather than scientific recommendations, attributed to limited research information on optimal fertilizer dosages. Therefore, education on the best fertilizer types and application rates for oil palm production is crucial for enhancing farmers’ knowledge, increasing yields per unit area, and boosting farmer incomes and national agricultural productivity.

4.4. Weeding

Weeds are harmful and unwanted plant species that reduce crop growth performance and yields in a field, including oil palm plantations. Weeds interfere with other agricultural operations, such as fertilization, harvesting, pest control, and disease control [34,35]. They indirectly affect crops by harboring parasitic plants, pathogens, and insect pests and eventually reducing the value of the cropland. In addition, weeds decrease the yields of a plantation, increase the cost of production, and expose an oil palm field to fire [36].

In the current study, 96% of the respondents control weeds by hand hoeing (93.1%), herbicides (0.3%), burning (0.8%), and cover crops (0.3%). However, weeds were removed only once in oil palm fields intercropped with other crops. Weeding once yearly is less effective than four times, as reported in other plantations [37]. Inefficient weeding is attributed to a lack of awareness of good management practices, the high cost of inputs, the lack of capital, and the prioritization of oil palm cultivation over other food crops. Since the 2020/21 season, the oil palm sector in Tanzania has grown remarkably due to support from the government through TARI and the private sector on the input supply chain. The distribution of improved seedlings, training on good agronomic practices, and the dissemination of related technologies to farmers and extension officers by TARI has recently engaged many new farmers, along with improvements in oil palm production knowledge.

The expansion of new farms is accelerating the commercial mono-cropping farming system, and farmers have to adopt an integrated eco-friendly approach for weed control, especially in young trees, to prevent growth inhibition and reduced yields. Iddris, Formaglio [35], reported productivity that was 50% higher in large-scale plantations compared to smallholder-owned farms, which was primarily driven by fertilization and weed control. More studies on identifying weed species density, composition and population changes, community dynamics, species diversity, and the efficacy of different control measures will lead to appropriate management decisions.

4.5. Intercropping in Oil Palm

Apart from providing farmers with additional food and income, intercropping promotes multiple livelihood options that benefit smallholder farmers and other stakeholders [38]. Crops such as beans, maize, sweet potatoes, banana, soyabean, peas, and groundnuts do not compete when planted between rows of young oil palm plants. At the early development stage, usually 4 to 5 years after planting, intercropping reduces the cost of weeding; improves soil fertility through nitrogen fixation when intercropped with legumes; and improves nutritional requirements, biodiversity, and sustainability and resilience for price fluctuations of fresh fruit bunches (FFBs) [38]. However, when intercropping with perennial crops, planting densities and crop integrations/combinations must be carefully considered to minimize the competition effect, which can lead to reduced oil palm yields. The general observations from the current intercrops revealed that farmers intercrop without technical know-how. This was confirmed by some farmers who intercrop cassava and pineapples in their oil palm field without a proper fertilization program for potassium replacement. It has been reported that intercropping of cassava and pineapples with oil palm is not profitable, unless there is an appropriate fertilization program that would ensure the replacement of potassium, which is constantly under competition with oil palm with such intercrops [39,40,41]. Dhandapani, Girkin [42], reported that intercropping reduces the negative environmental impacts of monoculture oil palm production. However, reduced yields were reported due to competition for nutrients, water, and light [43]. The current analysis indicated that farmers’ need to be educated on the advantages and disadvantages of intercropping crops and the choice of relevant intercrops in oil palm fields. The training of farmers on intercropping technologies that increase income, resilience, and biodiversity is required.

4.6. Access to Extension Services

An efficient extension service is the best approach to accelerate agricultural technologies and knowledge transfer from research to smallholder farmers [44,45,46,47]. In Tanzania, the government offers extension services through the District Agricultural, Livestock and Fishery Officer (DALFO). One extension officer coordinates and delivers services in one ward, an administrative sub-unit within a district comprising two or three villages. The current average ratio of extension officers to farmers is 1:2000. Reportedly, government extension officers represent only 10% of farming households in Tanzania. In addition, an extension officer works on different crops with a biased priority, depending on the district’s priority. Most of the interviewed extension officers were new and inexperienced and had an inadequate understanding of good agronomic practices in oil palm production. Before 2018, oil palm in Tanzania was not a priority crop; it was traditionally cultivated and not even taught in schools, colleges, and universities. In this regard, the government should adopt a new teaching curriculum to prepare teachers in schools, colleges, and universities for oil palm production and improvement. Basaruddin, Kannan [47], suggested that an extension officer needs to be well-trained and equipped with proper knowledge and skills. Extension officers should be trained occasionally to update them on modern oil palm farming and technologies, so that they can act as trainers and advisors of farmers in their working areas. Cloete, Bahta [48], stated that the success of extension workers needs clear goals, principles, teaching methods, and teaching tools for smallholder farmers. Similarly, Basaruddin, Kannan [47], reported that an ineffective extension system resulted from an inadequate number of extension workers, additional assigned duties, poor coordination and supervision of farmers, and a lack of knowledge of extension officers. Extension officers need to improve their skills and knowledge on oil palm production through access to training programs organized by TARI and partners and by networking with smallholder farmers through diverse teaching methods and time dedication.

4.7. Farmers’ Perceptions of Types of Oil Palm Genotypes Grown

Across the study area, farmers acknowledged the importance of the Tenera hybrid. Among the interviewed farmers, 97.4%, 54.6%, and 2% planted Dura, Tenera, and Pisifera, respectively. Dura has a thick shell (2–8 mm) and thin mesocarp (35–55%) with a low oil content, whereas Pisifera is shell-less with a 95% mesocarp. Despite having a high mesocarp content, Pisifera does not produce bunches and may bear fruits with a low oil-to-bunch ratio compared to Tenera [25]. Pisifera is female-sterile or semi-fertile with varying degrees of sterility but is used as a male parent in Tenera hybrid seed production, since its male inflorescence produces viable pollens [1,25]. Tenera has a thin shell (0.5–4 mm), a thick mesocarp (50–96%), and a high oil content and is the only form of oil palm fruits used for commercial planting. The thick-shell in the Dura genotype is controlled by the dominant homozygote gene (sh+sh+), whereas the shell-less property in Pisifera is controlled by the recessive homozygote gene (sh−sh−). The cross between Dura and Pisifera (D × P) results in a heterozygote Tenera hybrid (sh+sh−) with a thin-shell and a thick mesocarp [1,25].

Generally, access to improved Tenera in the study area was limited, suggesting the need to disseminate an improved hybrid to replace Dura. Many farmers reported the need for a Tenera hybrid for their field expansion and for replacing old trees planted in the 1970s. However, few farmers claimed that improved Tenera hybrid planting materials have a shorter life span than their locally grown Dura.

4.8. Farmers Preferred Traits of Oil Palm

Farmers in the study area preferred oil palm trees with high oil contents and many bunches. Goh, Mahamooth [25], reported high oil yields and many bunches per tress, high bunch weights, and the oil-to-bunch ratio as breeders’ traits of preference. These traits were reported to be controlled by many minor genes. Goh, Mahamooth [25], proposed that recurrent selection, like modified or reciprocal recurrent selection (RRS), is a useful method for improving oil palm. Conversely, farmers preferred short-stem oil palm to reduce labor costs during pruning and harvesting. Farmers in Kalenge and Kandaga in Uvinza claimed oil palm harvesting costs to be charged based on plant height, where harvesting from taller trees was expensive than shorter trees. Somyong, Walayaporn [49], proposed an oil palm breeding program focusing on short varieties that may reduce harvesting costs. Other farmers’ preferred traits were a tolerance to droughts and diseases, such as ganoderma, basal stem rot, and Fusarium wilt. These diseases were primarily reported in Kigoma Rural and Kigoma Urban. The importance of considering farmers’ traits of preferences has been reported in earlier studies. Mrema, Shimelis [50], and Kagimbo, Shimelis [51], proposed that any crop improvement program should consider the farmers’ preferred traits in a newly developed variety for a high adoption rate and food security. Kagimbo, Shimelis [51], pinpointed that in some cases in Africa, farmers reject varieties with superlative agronomic performance, if a newly released variety lacks their traits of interest. Therefore, developing an improved Tenera genotype that is resilient to climate change with farmers’ preferred traits is key for increased adoption in the study area.

4.9. Oil Palm Production Constraints

The oil palm industry involves different actors in production, processing, and marketing. However, this study focused on production challenges, opportunities, and farmers’ perceptions and preferred traits of oil palm. In this study, the main challenges along the oil palm value chain that affected production included an inadequate supply of improved seedlings (82.7%) and poor access to capital/credit (72.4%), for financing oil palm production was particularly significant, as they directly impacted farmers’ ability to improve productivity and invest in more efficient farming practices. Other challenges included are high input costs (59.4%); unreliable markets (56.4%); and the prevalence of pest and disease infestation (53.6%), especially rhinoceros beetles (rhinoceros oryctes) and Ganoderma (Ganoderma boninense), respectively, which affects most oil palm plants. Previous studies have reported similar production challenges and have recommended improvements for achieving sustainable production [52,53]. These findings imply that enabling access to financial services for oil palm farmers and access to improved seeds/seedlings will increase the efficiency of the oil palm industry in the country. Therefore, interventions should target developing improved oil palm planting material with a tolerance to pests and diseases for farmers and improving access to financial services relevant to oil palm farmers.

5. Conclusions

Sustainable oil palm production in Tanzania is hindered by multiple challenges, including an inadequate supply of improved planting materials, poor access to credit, a high cost of production inputs, poor market access, insect pests and diseases, poor production technologies, and knowledge gaps, that are the main reasons affecting oil palm production and productivity. Furthermore, constraints such as the limited use of fertilizers, the use of improved planting materials, the low frequency of weeding, inadequate extension services, intercropping that does not follow proper crop combinations, poor methods of weed control, and the lack of farmers’ preferred traits further impact oil palm production. The government should establish supportive policies to strengthen both public and private sector participation in the oil palm sector, ensuring better coordination at the local, national, and international level. Currently, TARI Kihinga is working to provide improved Tenera hybrid seeds, but a more systemic breeding program is necessary to enhance yields and meet farmers’ needs. Moreover, training programs for smallholder farmers and extension officers on good agricultural practices and the establishment of a formal oil palm seed system will promote the use of improved varieties and will support farm expansions for the economic development of Tanzania through local palm oil palm production and import substitution.