Jeremy’s latest Eat This Newsletter does “miracle wheat,” blue mangoes, and coffee. Quite the snack. Check it out. And if the blue mango thing sounds vaguely familiar, we Nibbled it here a few weeks back.
A seed of doubt
Kew’s Millennium Seed Bank has soft-launched SeedPOD, the Seed Portal for Online Data, a platform intended to make information about native seed collections, their availability and their propagation easier to find. The ambition is to tackle one of the persistent bottlenecks in ecological restoration: getting the right seed, from the right source, to the right place.
It is an interesting development, and one that fits into a broader restoration workflow I explored in an earlier post on restoration and digital tools. There is another tool that could usefully complement it: the RHS Garden Climate Assessment Tool (GCAT).
The idea is to connect three stages of a practical restoration exercise. First, identify a native species appropriate to the planting location. Then find out whether seed is available from the Millennium Seed Bank or another seed supplier. Finally, check whether the species is likely to remain climatically suitable at the planting site under future climate scenarios. If all goes well, obtain the seed and grow the plants.
I decided to try this with Markhamia obtusifolia, an indigenous African tree that might be an interesting addition to my garden in Nairobi.
The first stop was SeedPOD. Here I found a bit of a discrepancy. The older Millennium Seed Bank seed list website listed three accessions: one from Malawi and two from Botswana. The new SeedPOD interface showed just one accession, held by IIAM (Instituto de Investigação Agrária de Moçambique) in Maputo, and apparently the only one currently available. It is not clear whether this reflects a change in the underlying holdings, a difference in what the two databases display, or simply a difference in how availability is recorded. Either way, it raises a question about how complete and up to date the new interface is.
There was a little more information from BGCI’s PlantSearch, accessible through SeedPOD, which identified 15 ex situ sites holding the species. That suggests that the tree is represented in a reasonable number of living collections, although it does not tell us whether any of them can supply seed. BTW, the links from SeedPOD to GBIF and Genesys did not work when I tried them, and I could not find the promised list of seed suppliers either. So the next step in the workflow — actually finding some seed — remains elusive. (ChatGPT could not find any seed suppliers in Nairobi though it did flag a couple of plausible options for M. lutea.)
At least the climate assessment was more straightforward. Using GCAT, I was able to establish that M. obtusifolia should be climatically suitable for planting in Nairobi under most climate change scenarios. This is a useful check, although it does not settle questions of soil, provenance, or other ecological requirements. Climate suitability is not quite the same thing as ecological suitability.
None of this diminishes the value of SeedPOD’s underlying ambition. Bringing together information previously dispersed information is worthwhile. But the real test is whether a user can move easily and quickly from a species name to a viable source of seed, assess its suitability for a particular restoration site, and eventually get plants into the ground.
My little experiment has not quite completed that journey. SeedPOD helped identify ex situ holdings, and GCAT provided a positive climate-suitability assessment. But the route from a database record to an actual packet of seed is still surprisingly difficult to navigate. A platform intended to connect seed conservation with ecological restoration will ultimately be judged not just by how much information it brings together, but by whether people can use it to get seed into the ground.
A busy October, but what changed for opportunity crops?
The newsletter of the Community of Practice on Africa Forgotten and Underutilized Food Commodities just recycled a digest of posts from the blog of PAEPARD (Platform for African – European Partnership in Agricultural Research for Development) which included a plethora of events about opportunity crops that took place this October.
- AgrEcoNUS+ Launch Conference. The EU-funded project “Enhancing value chains and markets of neglected and underutilized crop species to contribute to agroecological transition in Africa” was launched on 1 October.
- African Union Inaugural Africa Seed Summit. Held on 5–7 October 2026 in Eswatini under the theme “Resilient Seed Systems for a Food Secure Africa.” We have mentioned it before.
- World Food Map: An Interactive Atlas of Edible Biodiversity. Launched online on 20 October 2026, it “brings together the foods people eat today with opportunity species that could diversify them, showing how diets can be improved with locally available foods that are culturally relevant, climate resilient, and nutritionally rich.”
- Advancing Underutilized Legumes for Food Security, Nutrition and Climate Resilience in Africa. The Society for Underutilized Legumes organised its 9th International Scientific Conference on 21–22 October, in Zaria, Nigeria.
- Bringing Forgotten Foods Back to the Table: Wageningen Symposium on Neglected and Underutilized Species. Wageningen University & Research brought together scientists, chefs and food-system specialists on 26 October to explore how forgotten foods can contribute to nutrition, biodiversity and culinary innovation.
- 16th Indigenous and Traditional Seed and Food Festival. Organised by PELUM Uganda, on 29–30 October, in Kampala, Uganda, the festival celebrated indigenous seeds, traditional foods, farmer-managed seed systems and agroecological knowledge for climate resilience.
Clearly a busy month for the NUS (neglected and underutilized species, aka opportunity crops) community. But what does it all mean? What do we know on 1 November that we didn’t know on 30 September that will help farmers to grow NUS? I just hope someone is keeping score of the results of all these meetings, and will let the rest of us know.
Brainfood: iNaturalist, Agroforestry mapping, Crop mapping, Intercrop mapping, Smartphone phenotyping, Bean phenology, Rice phenology, Grape database
- Guidelines and best practices for the scientific use of global iNaturalist data. Citizen scientists have generated an enormous photographic record of biodiversity, but turning those observations into reliable scientific data requires some care. Here’s some practical guidance on accessing and using iNaturalist records and images, addressing taxonomic uncertainty, geoprivacy, observer bias, attribution and reproducibility.
- Mapping the spatial extent of agroforestry lands across Africa by characterizing tree spatial arrangements. Moving from individual biodiversity observations to the landscape, this study uses remote sensing to identify agroforestry across Africa by looking at the spatial arrangement of trees.
- AI tool for mapping smallholder crops proves itself in Senegal. Once we know where trees and agricultural landscapes are, the next question is what is actually being grown. Here researchers use Tessera, an open-source AI model trained on satellite imagery, to identify different crops in smallholder systems, achieving 84% accuracy while requiring considerably less computing power and labelled training data than other approaches.
- National Mapping of Monoculture and Intercropped Maize via Phenology-Assisted Sampling: Application in Kenya. Mapping crops is one thing; distinguishing how they are grown is another. This study distinguished intercropped from monocropped maize in smallholder farming systems, using satellite time series and phenological information to generate training samples for machine-learning classification.
- Smartphone-based proximal sensing as an accessible platform for improved physiological phenotyping. From mapping whole agricultural landscapes we move to the research station, and from satellites to smartphones. Using a smartphone to capture repeated images of plants in cowpea plots generated high-resolution time series, from which could be extracted information on vegetative growth, flowering and pod production.
- Determination of dry bean physiological maturity from UAS RGB time series using a stacking ensemble of color-derived spectral features. The cowpea study shows what can be extracted from repeated close-range images of individual plants. Here the emphasis shifts to using drone-mounted cameras to estimate a specific breeding trait: physiological maturity in dry beans. Combining colour-based indicators with time-series features and machine learning achieved a prediction error of only 2.68 days.
- Multi-scale spatial-temporal remote sensing fusion for phenology identification in rice germplasm resources. If maturity can be estimated from drone imagery in dry beans, can similar approaches monitor hundreds of different rice genotypes in a breeding collection? Apparently yes, by combining frequent medium-resolution images with less frequent high-resolution UAV imagery.
- Digitalization of the genotypic and phenotypic information retained within the FEM grapevine germplasm bank. We end where phenotyping meets conservation. The Fondazione Edmund Mach grapevine collection brings together more than 3,000 accessions, with genetic identity assessed using microsatellite markers and phenotypic characterisation covering vine development, ampelographic traits and responses to biotic stresses. The paper describes the digitisation of this information in FEMVitisDB, a FAIR-compliant database with links to international data infrastructures. And not a drone in sight. Yet
Having a go at Crops·To·Go
The International Potato Centre (CIP) seems to have a new, AI-powered (of course) interface for requesting material from its genebank. They call it Crops·To·Go.
You have to register even if all you want to do is browse the collection. It didn’t take me very long to do so, but still: why? Requiring users to create an account before they can explore what is available seems an unnecessary barrier.
I was also not entirely convinced that all the search filters were working as expected. In particular, as I added evaluation-data filters, the number of accessions that purported to match them remained unchanged. It wasn’t clear whether the filters had been applied, whether they had returned no additional matches, or whether the matching count was simply not updating. Some more visible feedback on the effect of applying filters would make the experience a bit more reassuring.
Overall, this looks like a promising step towards making CIP’s germplasm more accessible, but the browsing experience and the transparency of the filtering process could do with some attention.