Integrating climate intelligence into digital agroadvisory systems: A scalable multi model framework for AgWise Engine
Agricultural production, particularly in smallholder systems, faces increasing risks from climate hazards such as floods, droughts, and typhoons, which directly threaten livelihoods. Timely access to climate advisories is paramount for strengthening adaptation strategies and minimizing climate chang...
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Informe técnico |
| Lenguaje: | Inglés |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://hdl.handle.net/10568/179597 |
| _version_ | 1855543696133455872 |
|---|---|
| author | Tibebe, Degefie Seid, Jemal Abera, Wuletawu Saito, Kazuki Mkuhlani, Siyabusa Leroux, Louise Mabilangan, Abigail Ghosh, Aniruddha Kihara, Job |
| author_browse | Abera, Wuletawu Ghosh, Aniruddha Kihara, Job Leroux, Louise Mabilangan, Abigail Mkuhlani, Siyabusa Saito, Kazuki Seid, Jemal Tibebe, Degefie |
| author_facet | Tibebe, Degefie Seid, Jemal Abera, Wuletawu Saito, Kazuki Mkuhlani, Siyabusa Leroux, Louise Mabilangan, Abigail Ghosh, Aniruddha Kihara, Job |
| author_sort | Tibebe, Degefie |
| collection | Repository of Agricultural Research Outputs (CGSpace) |
| description | Agricultural production, particularly in smallholder systems, faces increasing risks from climate hazards such as floods, droughts, and typhoons, which directly threaten livelihoods. Timely access to climate advisories is paramount for strengthening adaptation strategies and minimizing climate change impacts. However, a significant systemic challenge persists, widely recognized as the "climate information application gap". This gap exists because valuable climate forecast data, often produced by sophisticated meteorological agencies, fails to translate effectively into timely, location-specific, and actionable farm-level management decisions (Seid et al., 2020). Historically, agricultural extension services have relied on generic, often outdated, fertilizer recommendations derived from limited field trials conducted decades ago. This static, "blanket" recommendations fail to account for the crucial spatial heterogeneity of soils and the immense temporal variability inherent in modern seasonal rainfall patterns (AFR-2314, 2023) The resulting disconnect undermines farmer resilience and resource efficiency. The strategic goal for the AgWise1 Climate Intelligence (CI) integration is to engineer a fundamental paradigm shift: climate predictions must evolve from passive informational products into active decision-input variables that directly parameterize and modulate underlying agronomic optimization algorithms. Achieving genuine climate resilience necessitates a dual-timeframe strategy, a core lesson learned from successful implementations like WeRise, Climate+ and EDACaP-NextGen (Seid et al., 2023). This strategy requires integrating climate information across two distinct temporal scales: (i) Strategic Planning (3-6 Months): Utilizing seasonal forecasts to guide critical long-term decisions, including optimal variety selection, defining the best sowing or planting windows, and facilitating access to bundled financial services (AFR-2419, 2024) (ii) Tactical Execution (10-Day): Employing dynamic, high-frequency sub-seasonal forecasts to provide precise, in-season adjustments necessary for critical management actions, such as fine-tuning fertilizer application schedules, managing supplementary irrigation, and determining optimal harvest timing(CGIAR, 2024). The proposed framework for AgWise CI is structurally designed to overcome two specific technical challenges identified across foundational projects: (1) the translation problem converting complex, probabilistic climate outputs into agronomically meaningful parameters; (2) the integration problem embedding these climate-conditioned algorithms seamlessly within existing digital advisory architectures. |
| format | Informe técnico |
| id | CGSpace179597 |
| institution | CGIAR Consortium |
| language | Inglés |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| record_format | dspace |
| spelling | CGSpace1795972026-01-10T02:17:45Z Integrating climate intelligence into digital agroadvisory systems: A scalable multi model framework for AgWise Engine Tibebe, Degefie Seid, Jemal Abera, Wuletawu Saito, Kazuki Mkuhlani, Siyabusa Leroux, Louise Mabilangan, Abigail Ghosh, Aniruddha Kihara, Job agronomy climate action digital agriculture advisory services technology Agricultural production, particularly in smallholder systems, faces increasing risks from climate hazards such as floods, droughts, and typhoons, which directly threaten livelihoods. Timely access to climate advisories is paramount for strengthening adaptation strategies and minimizing climate change impacts. However, a significant systemic challenge persists, widely recognized as the "climate information application gap". This gap exists because valuable climate forecast data, often produced by sophisticated meteorological agencies, fails to translate effectively into timely, location-specific, and actionable farm-level management decisions (Seid et al., 2020). Historically, agricultural extension services have relied on generic, often outdated, fertilizer recommendations derived from limited field trials conducted decades ago. This static, "blanket" recommendations fail to account for the crucial spatial heterogeneity of soils and the immense temporal variability inherent in modern seasonal rainfall patterns (AFR-2314, 2023) The resulting disconnect undermines farmer resilience and resource efficiency. The strategic goal for the AgWise1 Climate Intelligence (CI) integration is to engineer a fundamental paradigm shift: climate predictions must evolve from passive informational products into active decision-input variables that directly parameterize and modulate underlying agronomic optimization algorithms. Achieving genuine climate resilience necessitates a dual-timeframe strategy, a core lesson learned from successful implementations like WeRise, Climate+ and EDACaP-NextGen (Seid et al., 2023). This strategy requires integrating climate information across two distinct temporal scales: (i) Strategic Planning (3-6 Months): Utilizing seasonal forecasts to guide critical long-term decisions, including optimal variety selection, defining the best sowing or planting windows, and facilitating access to bundled financial services (AFR-2419, 2024) (ii) Tactical Execution (10-Day): Employing dynamic, high-frequency sub-seasonal forecasts to provide precise, in-season adjustments necessary for critical management actions, such as fine-tuning fertilizer application schedules, managing supplementary irrigation, and determining optimal harvest timing(CGIAR, 2024). The proposed framework for AgWise CI is structurally designed to overcome two specific technical challenges identified across foundational projects: (1) the translation problem converting complex, probabilistic climate outputs into agronomically meaningful parameters; (2) the integration problem embedding these climate-conditioned algorithms seamlessly within existing digital advisory architectures. 2025-12-29 2026-01-09T13:33:58Z 2026-01-09T13:33:58Z Report https://hdl.handle.net/10568/179597 en Open Access application/pdf Tibebe, D.; Seid, J.; Abera, W.; Saito, K.; Mkuhlani, S.; Leroux, L.; Mabilangan, A.; Ghosh, A.; Kihara, J. (2025) Integrating climate intelligence into digital agroadvisory systems: A scalable multi model framework for AgWise Engine. CGIAR Sustainable Farming Science Program. 21 p. |
| spellingShingle | agronomy climate action digital agriculture advisory services technology Tibebe, Degefie Seid, Jemal Abera, Wuletawu Saito, Kazuki Mkuhlani, Siyabusa Leroux, Louise Mabilangan, Abigail Ghosh, Aniruddha Kihara, Job Integrating climate intelligence into digital agroadvisory systems: A scalable multi model framework for AgWise Engine |
| title | Integrating climate intelligence into digital agroadvisory systems: A scalable multi model framework for AgWise Engine |
| title_full | Integrating climate intelligence into digital agroadvisory systems: A scalable multi model framework for AgWise Engine |
| title_fullStr | Integrating climate intelligence into digital agroadvisory systems: A scalable multi model framework for AgWise Engine |
| title_full_unstemmed | Integrating climate intelligence into digital agroadvisory systems: A scalable multi model framework for AgWise Engine |
| title_short | Integrating climate intelligence into digital agroadvisory systems: A scalable multi model framework for AgWise Engine |
| title_sort | integrating climate intelligence into digital agroadvisory systems a scalable multi model framework for agwise engine |
| topic | agronomy climate action digital agriculture advisory services technology |
| url | https://hdl.handle.net/10568/179597 |
| work_keys_str_mv | AT tibebedegefie integratingclimateintelligenceintodigitalagroadvisorysystemsascalablemultimodelframeworkforagwiseengine AT seidjemal integratingclimateintelligenceintodigitalagroadvisorysystemsascalablemultimodelframeworkforagwiseengine AT aberawuletawu integratingclimateintelligenceintodigitalagroadvisorysystemsascalablemultimodelframeworkforagwiseengine AT saitokazuki integratingclimateintelligenceintodigitalagroadvisorysystemsascalablemultimodelframeworkforagwiseengine AT mkuhlanisiyabusa integratingclimateintelligenceintodigitalagroadvisorysystemsascalablemultimodelframeworkforagwiseengine AT lerouxlouise integratingclimateintelligenceintodigitalagroadvisorysystemsascalablemultimodelframeworkforagwiseengine AT mabilanganabigail integratingclimateintelligenceintodigitalagroadvisorysystemsascalablemultimodelframeworkforagwiseengine AT ghoshaniruddha integratingclimateintelligenceintodigitalagroadvisorysystemsascalablemultimodelframeworkforagwiseengine AT kiharajob integratingclimateintelligenceintodigitalagroadvisorysystemsascalablemultimodelframeworkforagwiseengine |