Tuesday, July 09, 2024

PLANTOLOGY

Archaeologists report earliest evidence for plant farming in east Africa



WASHINGTON UNIVERSITY IN ST. LOUIS

Kakapel 

IMAGE: 

LOCATED IN THE FOOTHILLS OF MOUNT ELGON NEAR THE KENYA-UGANDA BORDER, KAKAPEL ROCKSHELTER IS THE SITE WHERE WASHU ARCHAEOLOGIST NATALIE MUELLER AND HER COLLABORATORS HAVE UNCOVERED THE EARLIEST EVIDENCE FOR PLANT FARMING IN EAST AFRICA. 

view more 

CREDIT: STEVEN GOLDSTEIN




A trove of ancient plant remains excavated in Kenya helps explain the history of plant farming in equatorial eastern Africa, a region long thought to be important for early farming but where scant evidence from actual physical crops has been previously uncovered.

In a new study published July 10 in the Proceedings of the Royal Society B, archaeologists from Washington University in St. Louis, the University of Pittsburgh and their colleagues report the largest and most extensively dated archaeobotanical record from interior east Africa.

Up until now, scientists have had virtually no success in gathering ancient plant remains from east Africa and, as a result, have had little idea where and how early plant farming got its start in the large and diverse area comprising Kenya, Tanzania and Uganda.

“There are many narratives about how agriculture began in east Africa, but there’s not a lot of direct evidence of the plants themselves,” said WashU’s Natalie Mueller, an assistant professor of archaeology in Arts & Sciences and co-first author of the new study. The work was conducted at the Kakapel Rockshelter in the Lake Victoria region of Kenya.

“We found a huge assemblage of plants, including a lot of crop remains,” Mueller said. “The past shows a rich history of diverse and flexible farming systems in the region, in opposition to modern stereotypes about Africa.”

The new research reveals a pattern of gradual introductions of different crops that originated from different parts of Africa.

In particular, the remnants of cowpea discovered at Kakapel rock shelter and directly dated to 2,300 years ago constitute the earliest documented arrival of a domesticated crop — and presumably of farming lifeways — to eastern Africa. Cowpea is assumed to have originated in west Africa and to have arrived in the Lake Victoria basin concurrent with the spread of Bantu-speaking peoples migrating from central Africa, the study authors said.

“Our findings at Kakapel reveal the earliest evidence of domesticated crops in east Africa, reflecting the dynamic interactions between local herders and incoming Bantu-speaking farmers,” said Emmanuel Ndiema from the National Museums of Kenya, a project partner. “This study exemplifies National Museums of Kenya’s commitment to uncovering the deep historical roots of Kenya’s agricultural heritage and fostering an appreciation of how past human adaptations can inform future food security and environmental sustainability.”

Constantly changing landscape

Situated north of Lake Victoria, in the foothills of Mount Elgon near the Kenya-Uganda border, Kakapel is a recognized rock art site that contains archaeological artifacts that reflect more than 9,000 years of human occupation in the region. The site has been recognized as a Kenyan national monument since 2004.

“Kakapel Rockshelter is one of the only sites in the region where we can see such a long sequence of occupation by so many diverse communities,” said Steven T. Goldstein, an anthropological archaeologist at the University of Pittsburgh (WashU PhD ’17), the other first author of this study. “Using our innovative approaches to excavation, we have been uniquely able to detect the arrival of domesticated plants and animals into Kenya and study the impacts of these introductions on local environments, human technology and sociocultural systems.”

Mueller first joined Goldstein and National Museums of Kenya to conduct excavations at the Kakapel Rockshelter site in 2018. Their work is ongoing. Mueller is the lead scientist for plant investigations at Kakapel; the Max Planck Institute of Geoanthropology (in Jena, Germany) is another partner on the project.

Mueller used a flotation technique to separate remnants of wild and domesticated plant species from ashes and other debris in a hearth excavated at Kakapel. Although she has used this technique in her research in many other parts of the world, it is sometimes difficult to use this approach in water-scarce locations — so it has not been widely used in east Africa.

The scientists used direct radiocarbon dating on carbonized seeds to document the arrival of cowpea (also known as the black-eyed pea, today an important legume around the world) about 2,300 years ago, at about the same time that people in this area began to use domesticated cattle. Researchers also found evidence that sorghum arrived from the northeast at least 1,000 years ago. They also recovered hundreds of finger millet seeds, dating back to at least 1,000 years ago. This crop is indigenous to eastern Africa and is an important heritage crop for the communities that live near Kakapel today.

One unusual crop that Mueller uncovered was field pea (Pisum), burnt but perfectly intact. Peas were not previously considered to be part of early agriculture in this region. “To our knowledge, this is the only evidence of peas in Iron Age eastern Africa,” Mueller said.

The exceptional pea is pictured in the paper, and it represents its own little mystery. “The standard peas that we eat in North America were domesticated in the near east,” Mueller said. “They were grown in Egypt and probably ended up in east Africa by traveling down the Nile through Sudan, which is also likely how sorghum ended up in east Africa. But there is another kind of pea that was domesticated independently in Ethiopia called the Abyssinian pea, and our sample could be either one!”

Many of the plant remnants that Mueller and her team found at Kakapel could not be positively identified, Mueller said, because even modern scientists working in Kenya, Tanzania and Uganda today don’t have access to a good reference collection of samples of plants from east Africa. (As a separate project, Mueller is currently working on building such a comparative collection of Tanzania’s plants.)

“Our work shows that African farming was constantly changing as people migrated, adopted new crops and abandoned others at a local level,” Mueller said. “Prior to European colonialism, community-scale flexibility and decision-making was critical for food security — and it still is in many places.”

Findings from this study may have implications for many other fields, Mueller said, including historical linguistics, plant science and genetics, African history and domestication studies.

Mueller is continuing to work on identifying the wild plants in the assemblage, especially those from the oldest parts of the site, before the beginning of agriculture. “This is where human evolution occurred,” Mueller said. “This is where hunting and gathering was invented by people at the dawn of time. But there has been no archaeological evidence about which plants hunter-gatherers were eating from this region. If we can get that kind of information from this assemblage, then that is a great contribution.”

One unusual crop that Mueller uncovered was field pea, burnt but perfectly intact. Peas were not previously considered to be part of early agriculture in this region. 

CREDIT

Courtesy of Proc. Royal Soc. B


 How a plant app helps identify the consequences of climate change


By leveraging millions of time-stamped observations, researchers can identify plant rhythms and ecological patterns year-round


Peer-Reviewed Publication

GERMAN CENTRE FOR INTEGRATIVE BIODIVERSITY RESEARCH (IDIV) HALLE-JENA-LEIPZIG

The Flora Incognita app makes it easy to identify plants with a smartphone. 

IMAGE: 

THE FLORA INCOGNITA APP MAKES IT EASY TO IDENTIFY PLANTS WITH A SMARTPHONE.

view more 

CREDIT: FLORA INCOGNITA



Plants are known to respond to seasonal changes by budding, leafing, and flowering. As climate change stands to shift these so-called phenological stages in the life cycle of plants, access to data about phenological changes – from many different locations and in different plants – can be used to draw conclusions about the actual effects of climate change. However, conducting such analyses require a large amount of data and data collection of this scale would be unthinkable without the help of citizen scientists. “The problem is that the quality of the data suffers when fewer people engage as citizen scientists and stop collecting data,” says first author Karin Mora, research fellow at Leipzig University and iDiv.

Mobile apps like Flora Incognita could help solve this issue. The app allows users to identify unknown wild plants within a matter of seconds. “When I take a picture of a plant with the app, the observation is recorded with the (exact) location as well as a time stamp,” explains co-author Jana Wäldchen from the Max Planck Institute for Biogeochemistry (MPI-BGC), who developed the app with colleagues from TU Ilmenau. “Millions of time-stamped plant observations from different regions have been collected by now.” Although satellite data also records the phenology of entire ecosystems from above, they do not provide information about the processes taking place on the ground.

Plants show synchronised response

The researchers developed an algorithm that draws on almost 10 million observations of nearly 3,000 plants species identified between 2018 and 2021 in Germany by users of Flora Incognita. The data show that each individual plant has its own cycle as to when it begins a flowering or growth phase. Furthermore, the scientists were able to show that group behaviour arises from the behaviour of individuals. From this, they were able to derive ecological patterns and investigate how these change with the seasons. For example, ecosystems by rivers differ from those in the mountains, where phenological events start later.

The algorithm also accounts for the observational tendencies of Flora Incognita users, whose data collection is far from systematic. For example, users record more observations on the weekend and in densely populated areas. “Our method can automatically isolate these effects from the ecological patterns,” Karin Mora explains. “Fewer observations don’t necessarily mean that we can’t record the synchronisation. Of course, there are very few observations in the middle of winter, but there are also very few plants that can be observed during that time.”

It is known that climate change is causing seasonal shifts – for example, spring is arriving earlier and earlier. How this affects the relationship between plants and pollinating insects and therefore potentially also food security is still being subject to further research. The new algorithm can now be used to better analyse the effects of these changes on the plant world.

 

This study was funded by the Deutsche Forschungsgemeinschaft (DFG; FZT-118) and the iDiv Flexpool.

LGNet revolutionizes plant disease detection for enhanced crop protection



NANJING AGRICULTURAL UNIVERSITY THE ACADEMY OF SCIENCE
Fig.5 

IMAGE: 

THE ACCURACY OF EACH EPOCH.

view more 

CREDIT: THE AUTHORS




A research team has developed LGNet, a dual-branch network that combines convolutional neural networks (CNNs) and visual transformers (VTs) for plant disease identification. LGNet effectively fuses local and global features, achieving state-of-the-art recognition accuracies of 88.74% on the AI Challenger 2018 dataset and 99.08% on the self-collected corn disease dataset. This innovative approach enhances disease sensing capabilities and offers the potential for the development of efficient and robust plant disease recognition models, which are crucial for improving agricultural production and ensuring crop safety in diverse environments.

Safeguarding agricultural production is vital for economic growth, as plant diseases significantly threaten crop yields. The traditional methods of identifying plant diseases, which rely  on the farmers' experience, are time-consuming and inadequate for large-scale cultivation. Recent advancements in image processing and deep learning have improved plant disease recognition, yet existing methods using only CNNs or VTs fall short due to their limited feature perception.

study (DOI: 10.34133/plantphenomics.0208) published in Plant Phenomics on 21 Jun 2024, proposes LGNet, a dual-branch network combining CNNs and VTs that enhances both local and global feature extraction, achieving state-of-the-art performance on major datasets.

The research divided LGNet's parameters into two parts for training, utilizing pretrained weights on ImageNet 1k for the dual-branch backbone network and fine-tuning with different learning rates. The model was optimized with SGD, momentum, and weight decay, and trained on a Windows 11 system with an NVIDIA GeForce RTX 3090 GPU and PyTorch. For evaluation purpose, cross-entropy loss was used, while online data augmentation enhanced generalization. LGNet's performance was compared to single models ConvNeXt-Tiny and Swin Transformer-Tiny. The initial training accuracies were high for all models, but LGNet's accuracy improved significantly, surpassing the others by 1-2%. On the AI Challenger 2018 and SCD datasets, LGNet achieved 88.74% and 99.08% accuracy, respectively, outperforming the single models. Ablation experiments showed that both the AFF and HMUFF modules enhanced performance, with the full LGNet model achieving the best results, demonstrating the effectiveness of the dual-branch network and feature fusion techniques.

According to the study's lead researcher, Xin Zhang, “The development of robust plant disease recognition models, and improving the generalization ability of these models in real-world environments, is highly important for agricultural production.”

In summary, this study presents LGNet, a dual-branch network combining CNNs and VTs for enhanced plant disease identification. Future research will focus on knowledge distillation to create lightweight, high-performance models for mobile deployment and on obtaining more real-world data to enhance model robustness, thereby improving precision agriculture and ensuring crop safety.

###

References

DOI

10.34133/plantphenomics.0208

Original Source URL

https://doi.org/10.34133/plantphenomics.0208

Funding information

This study was supported by the National Key Research and Development Program of China (2021YFE0113700), the National Natural Science Foundation of China (32360705;31960555), the Guizhou Provincial Science and Technology Program (2019-1410;HZJD[2022]001), the Outstanding Young Scientist Program of Guizhou Province (KY2021-026), and the Program for Introducing Talents to Chinese Universities (111 Program; D20023).

About Plant Phenomics

Plant Phenomics is an Open Access journal published in affiliation with the State Key Laboratory of Crop Genetics & Germplasm Enhancement, Nanjing Agricultural University (NAU) and published by the American Association for the Advancement of Science (AAAS). Like all partners participating in the Science Partner Journal program, Plant Phenomics is editorially independent from the Science family of journals. Editorial decisions and scientific activities pursued by the journal's Editorial Board are made independently, based on scientific merit and adhering to the highest standards for accurate and ethical promotion of science. These decisions and activities are in no way influenced by the financial support of NAU, NAU administration, or any other institutions and sponsors. The Editorial Board is solely responsible for all content published in the journal. To learn more about the Science Partner Journal program, visit the SPJ program homepage.

No comments: