Pest management science often disregards farming system complexities

The paper addresses how pest management science in the Global South aligns with the complexity of farming systems and whether it supports transformative, sustainable crop protection. Against the backdrop of significant pest-induced crop losses, escalating pesticide use and toxicity, and widespread environmental and health externalities, the authors posit that a paradigm shift towards agroecology and systems-based approaches is needed. They note that despite decades of Integrated Pest Management (IPM), pesticide use has not declined and that scientific agendas in developing countries have not been systematically charted. The study’s purpose is to quantitatively map the scope, maturity, and thematic focus of pest management research across 65 developing countries, examining target taxa, crops, research types, IPM themes, coverage of farming system variables, and inclusion of companion biota, to assess how well science incorporates system complexity and informs policy and practice.

The authors situate their work within a body of scholarship documenting: large global crop losses to pests; mounting environmental and health costs from synthetic pesticides; calls for agroecological, biodiversity-based pest management; and critiques that IPM has often failed to reduce pesticide use. Prior systematic reviews in Western contexts revealed conceptually skewed research agendas and gaps in pest management science, but comparable mapping for the Global South was lacking. The literature highlights the importance of system redesign, cross-disciplinary integration, and leveraging ecological processes (e.g., natural enemies, plant defenses) across field-to-landscape scales. It also underscores barriers such as disciplinary specialization, incentives, and limited decision-relevant tools that hinder uptake of non-chemical management.

The study used bibliometric and multi-method analyses focusing on 65 developing countries across four sub-regions (Southeast Asia, Latin America & the Caribbean, West Africa, Middle East) over 2010–2020. Using Web of Science (Core Collection) topic searches (TS = ((field OR crop*) AND (pest) AND country), where country was replaced by each focal nation), the authors retrieved publications conducted in or co-authored by researchers from those countries. Searches (Aug–Oct 2022) yielded 5,924 records, which were screened via titles and abstracts. Exclusions: human/animal/urban pests (except termites affecting crops), zoonotic/vector-borne disease vectors (e.g., mosquitoes), pesticide handling/PPE/residue/dissipation/degradation/analytical method validation studies. Studies on pest susceptibility or resistance to pesticides (lab, semi-field, field) and storage pests were retained. Duplicates were removed. The final corpus comprised 3,407 peer-reviewed publications. Each abstract was categorized for: focal herbivore taxa (up to six genus/species-level entries), focal crops (14 categories, building on FAO ICC), research type (laboratory/desktop; review; greenhouse/semi-field; field), IPM thematic areas (eight core themes plus five finer-resolution categories: host plant resistance (HPR), sterile insect technique (SIT), insecticide resistance management (IRM), botanical insecticides, decision thresholds), farming system variables (15 variables spanning soil, plant, field, farm, landscape, social strata; accounting for space, time, gene dimensions), and companion biota (six groups: weeds/non-crop plants; plant pathogens/diseases; non-pest herbivores; soil/rhizosphere biota; pollinators; biological control agents (BCAs), further divided into vertebrates, invertebrate predators, invertebrate parasitoids, microorganisms, viruses). For food crops, scientific attention was compared with their share in a global reference diet and with annual insecticide mass and total insecticide hazard load (HL = Σ[Mi/(NOAELi × 365)] using PEST-CHEMGRIDS application rates and NOAEL values from published sources). The incidence of insecticide resistance (IR) for the 100 most-studied herbivores was compiled to relate taxon-specific research attention to IR. Statistical analyses (normality/homoscedasticity checks, transformations as needed) included linear regression (attention vs IR incidence), Kruskal–Wallis tests (numbers of system variables and companion biota across sub-regions), and chi-square tests (geographical biases by taxon; differences across research types, IPM themes, system variables). Visualization included heatmaps of co-considered system variables and radar plots for relative coverage of system variables and companion biota. A focused stratified analysis grouped 13 system variables into six strata (soil, plant, field, farm, landscape, social) for five key pests (Bemisia tabaci, Spodoptera frugiperda, Tuta absoluta, Tetranychus urticae, Helicoverpa armigera), excluding the focal pest and management regime variables.

• Corpus and taxa coverage: 3,407 publications after screening (country-level output range 0–459 over 10 years). Across sub-regions, 881 herbivore taxa were covered in 2,891 instances. Single-taxon studies accounted for 57.4% of studies. Common pests included Bemisia tabaci (110 studies), Spodoptera frugiperda (94), Tuta absoluta (80), Tetranychus urticae (72), Helicoverpa armigera (67). 94.6% of herbivore taxa featured in fewer than one publication per year; only 0.6% received more than five publications per year. For 43% of herbivore taxa, studies covered biological control agents (BCAs); 95.4% of these featured in fewer than one publication per year. BCAs were most commonly addressed for T. absoluta (≈3.9 studies/year), S. frugiperda (≈3.2), and T. urticae (≈3.2). Scientific attention increased with IR incidence (linear regression: F ≈ 1.99e3, p < 0.01, R2 = 0.41), with exceptions for recent invasives (S. frugiperda, T. absoluta) and some IR-prone but understudied taxa (e.g., Spodoptera exigua, S. litura).
• Crops: Scientific effort was aligned with crop share in the global reference diet (Pearson r = 0.925, p < 0.01) but diverged from contributions to insecticide mass and hazard load (Spearman ρ ≈ 0.462, p = 0.13; ρ ≈ 0.315, p = 0.32). Attention concentrated on cereal grains (17.6% studies), fruits (17.3%), and non-starchy vegetables (15.1%).
• Research types and IPM themes: 47.9% involved laboratory/desktop research; 7.8% were reviews; 6.2% greenhouse/semi-field; 49.0% field (some studies covered multiple types). Among eight IPM thematic areas, bio-ecology, preventative non-chemical, and curative non-chemical management were most common. HPR appeared in 7.7% of studies; SIT 1.1%; decision thresholds 0.5%; botanical insecticides 5.9%; BCAs 32.5%; preventative chemical management 0.3%. Of 2,086 management-centered studies, mean tactics per study were 1.2 ± 0.5; 80.2% evaluated a single tactic; 28.6% involved curative chemical control. In insecticide studies, 22.2% evaluated non-target impacts on/compatibility with BCAs. Yield and revenue were endpoints in only 9.3% and 2.4% of studies, respectively.
• Farming system variables and companion biota: In 1,832 greenhouse and (semi-)field studies, an average of 1.8 ± 1.0 of 15 system variables and 0.6 ± 0.8 of 6 companion biota were considered. Frequently included variables: focal pest (81.1%), pest management regime (29.0%), crop genetics/phenology (21.0%), intercropping (space-based interspecific diversity, 6.5%). Rarely considered: crop rotation (time-based interspecific diversity, 1.4%), soil moisture/irrigation (0.9%), intraspecific plant diversity (0.2%). Companion biota coverage: BCAs (34.7%), non-crop plants (7.6%), plant pathogens (4.5%), soil fauna/flora (3.1%), pollinators (1.6%). BCA types included vertebrates (1.2%), invertebrate predators (16.5%), invertebrate parasitoids (15.9%), microorganisms (8.5%), viruses (1.0%).
• Effects of management type: Pest management type significantly affected the number of system variables (ANOVA p < 0.001) and the proportional coverage of strata (soil, plant, field, farm, landscape, social; all p < 0.001 by chi-square). Generally, studies emphasized plant and field strata; farm/landscape/social and below-ground processes were underrepresented.
• Five focal pests: Geographic emphases varied (e.g., T. urticae: Middle East; S. frugiperda: Latin America). For T. urticae, 77.8% of studies were laboratory/desktop; (semi-)field or greenhouse research made up 18.1% and 25.8% for T. urticae and S. frugiperda, respectively. Bio-ecology was the most common theme (37.5–55.9% across taxa). Curative measures (chemical and non-chemical) featured more than preventative non-chemical ones by 31.0%, 14.3%, 86.4%, 100.0%, and 19.0% for B. tabaci, S. frugiperda, T. absoluta, T. urticae, and H. armigera, respectively. BCA coverage differed among taxa (χ2 = 10.544, p = 0.032; 28.2% in B. tabaci up to 48.8% in T. absoluta). Botanical insecticides appeared in 6.3–13.9% of studies. In field-focused work, the focal pest variable appeared in 94.3–100% of studies; plant and field strata dominated (up to 42.8% and 21.6% of studies); social, farm, landscape, and soil strata were consistently less covered. Numbers of system variables (≈1.8–2.0) and companion biota (≈0.6–1.0) per field study did not differ significantly between taxa.
• Association with broader system research: Herbivore taxa for which less-common system variables or non-BCA companion biota were studied had higher overall research output (Kruskal–Wallis H = 220.178 and 158.062, both p < 0.001).

The analysis reveals that pest management science in 65 developing countries is largely reductionist, pest-centric, and focused on curative tactics within simplified experimental contexts. Nearly half of studies are conducted in laboratory settings; 80% of management studies assess a single tactic; and most field studies consider two or fewer farming system variables, with minimal attention to social, farm, or landscape strata and below-ground processes. Despite strong coverage of bio-ecology and interest in non-chemical options such as BCAs, research often treats these as isolated, commoditized tools rather than integrating multiple preventative measures across relevant spatial and temporal scales. Decision-support (e.g., thresholds) is rarely developed or validated, and endpoints relevant to farmers (yield, revenue) are infrequently reported, which likely constrains adoption. Taxonomic attention is skewed toward IR-prone and invasive herbivores, while nutrient-dense, pesticide-intensive crops (fruits, legumes, starchy vegetables) are under-studied. Cross-disciplinary integration with pathologists, pollination ecologists, soil biologists, and social scientists is limited, as shown by the low coverage of companion biota and social variables. Structural barriers—disciplinary specialization, funding and incentive systems, short project cycles—further impede systems-oriented, problem-driven research. Collectively, the findings suggest that the current scientific enterprise contributes little to holistic resilience thinking or truly integrated pest management and is unlikely to catalyze transformative change without reorientation toward systems approaches.

This study provides the first systematic, quantitative mapping of pest management science across 65 developing countries, showing pervasive simplification, isolated treatment of tactics, narrow taxonomic focus, and limited engagement with farming system complexity and decision-relevant metrics. To realize sustainable, resilient crop protection and reduce pesticide externalities, research agendas should: integrate multiple farming system variables across soil-to-landscape and social strata; emphasize preventative, biodiversity-based approaches (e.g., diversification in space, time, and genetics); co-develop decision thresholds and other tools with end-users; and strengthen cross-disciplinary collaborations (ecology, agronomy, pathology, pollination, soil science, and social-behavioral sciences). Adoption can be accelerated through landscape-level framings, new decision frameworks (e.g., biodiversity spiral, hierarchical stratification, integrative food web analytics), prioritized research portfolios, revised incentives and funding schemes, enabling policies, and participatory platforms (e.g., farmer–scientist co-learning alliances, farmer field schools). Aligning scientific practice with the complexity of farming systems is essential for tangible impacts on food security, biodiversity conservation, and human health.

The analysis focuses on indexed, English-language publications in Web of Science and excludes substantial outputs from some countries (e.g., Brazil, China, Western nations), potentially biasing coverage of cosmopolitan pests. Abstract-based categorization may omit nuances present only in full texts. Some research types overlapped within studies, and reported statistical values for certain analyses are summarized rather than exhaustively detailed. The study does not directly measure on-farm adoption or outcomes, limiting causal inference between research focus and practice.