Malherbe Megan, Webb Nicole, Palisson-Kramer Magdalena, Ndiema Emmanuel K, Braun David R, Haeusler Martin, Forrest Frances
Institute of Evolutionary Medicine, University of Zürich, 190 Winterthurerstrasse, Zürich, 8001, Switzerland; Human Evolution Research Institute, University of Cape Town, Woolsack Drive, Rondebosch, Cape Town, 7701, South Africa.
Institute of Evolutionary Medicine, University of Zürich, 190 Winterthurerstrasse, Zürich, 8001, Switzerland; Institute of Archaeological Sciences, Senckenberg Centre for Human Evolution and Palaeoenvironment, Eberhard Karls University of Tübingen, 23 Rümelinstrasse, Tübingen, 72074, Germany.
J Hum Evol. 2025 Jun;203:103681. doi: 10.1016/j.jhevol.2025.103681. Epub 2025 Apr 24.
This research presents a new method of ecological morphology (ecomorphology) analysis using three-dimensional geometric morphometrics to quantify shape variation in extant bovid metapodials with known habitat preferences. Extant data were used to create a model for classifying bones into distinct habitat categories and to test functional hypotheses related to locomotor behavior in different habitats. The model was then applied to fossils from the Koobi Fora Formation, Kenya, to assess the environmental context during important events in hominin evolution. The use of three-dimensional geometric morphometrics demonstrates significant improvement over traditional methods using caliper measurements. Discriminant function analysis successfully classified 94% of metacarpals and 93% of metatarsals into their correct habitat categories for modern specimens. The protocol was reduced to a subset of landmarks focused on the distal epiphyses. This model produced greater overlap, but classification success rates remained high, with 82% and 83% correct classification for modern metacarpals and metatarsals, respectively. We applied the reduced model to metapodials from Upper Burgi (1.98-1.87 Ma), KBS (1.87-1.56 Ma), and Okote (1.56-1.38 Ma) members in the Koobi Fora Formation. This location is important to understanding human evolution, fossil diversity, and paleoecology. Moreover, previous studies on faunal abundance, paleosol carbonates, and carbon isotopes provide a robust framework to compare the findings of this study. Our analyses classified the majority of fossil specimens as open-habitat dwellers, with a few specimens grouped as closed-adapted, the highest number of these falling within the Okote Member sample. This suggests that open and likely xeric environments dominated the East Turkana region during the Early Pleistocene. These findings are consistent with many previous reconstructions, though with a more open signal for the Okote Member than expected based on bovid abundance research.
本研究提出了一种新的生态形态学（ecomorphology）分析方法，该方法使用三维几何形态测量学来量化具有已知栖息地偏好的现存牛科动物掌骨和跖骨的形状变化。利用现存数据创建了一个模型，用于将骨骼分类到不同的栖息地类别，并检验与不同栖息地运动行为相关的功能假设。然后将该模型应用于肯尼亚科比福拉组的化石，以评估人类进化重要事件期间的环境背景。三维几何形态测量学的使用表明，相较于使用卡尺测量的传统方法有显著改进。判别函数分析成功地将94%的掌骨和93%的跖骨正确分类到现代标本的相应栖息地类别中。该方案简化为一组专注于远端骨骺的地标点。这个模型产生了更大的重叠，但分类成功率仍然很高，现代掌骨和跖骨的正确分类率分别为82%和83%。我们将简化后的模型应用于科比福拉组上部布尔吉（198 - 187万年前）、KBS（187 - 156万年前）和奥科特（156 - 138万年前）成员的掌骨和跖骨。这个地点对于理解人类进化、化石多样性和古生态学很重要。此外，先前关于动物丰度、古土壤碳酸盐和碳同位素的研究提供了一个有力的框架来比较本研究的结果。我们的分析将大多数化石标本归类为开阔栖息地的居住者，少数标本归类为适应封闭环境的，其中数量最多的属于奥科特成员样本。这表明在早更新世期间，开阔且可能干旱的环境主导了东图尔卡纳地区。这些发现与许多先前的重建结果一致，不过基于牛科动物丰度研究，奥科特成员的开阔信号比预期更强。
